Originally Posted by
ring gear
It has always been taught in my experience, that if landing on any deck/platform or surface that moves, turn the "automatics" off after landing and it is normally the last thing to engage before getting airborne again. The rationale being the fundamentals of the two different systems - SAS & ATT (or APLT). But each system has direct control on the tip path plane of the disc with the very real potential of taking heads off.
SAS on most, if not all helicopters is called an "inner" loop system. That is, it acts completely independently of your stick position. It is designed to provide short term dampening and gust alleviation from short term changes in pitch/roll/yaw rate sensors (gyros/accelerometers). This short term change is then processed (in the AFCS computer) and a signal sent to the hydraulic/electric actuators to command an input change to the disc. SAS actuators are the "muscle" to convey the desired disc change that the SAS portion of the AFCS desires. This means the SAS by itself can physical change the tip path plane without any input from the pilot and without any change to the pilot's stick position.....ie an "inner' loop control input. The pilot has no idea how much of an input the SAS is giving to the disc. The SAS normally only has a limited control authority - something like 5-15% only (don't quote me on this ...it has been some time and may vary from system to system). None-the-less, SAS has the power to change your disc's tip path plane without your input and subject to external forces beyond your control....sometimes significantly so depending upon the external disturbance.
ATT/APLT is called an "outer" loop system. It provides long term stability such as Attitude hold and long term navigation. As such, it takes inputs for not only the pitch/roll/yaw rate sensors but the attitude gyro, the navigation inputs, accelerometers, pressure instruments etc. The Autopilot ("long term") part of the AFCS computer calculates the Long Term stability or aircraft positional information eg heading, airspeed, height, ROC, etc and determines where it needs to position the flight controls to achieve that result. The AFCS/APLT does this by sending its processed signal as an output to the Force Trim, which then inputs to the disc via the Force Trim motors....ie the Force Trim is the "muscle" for the APLT to operate....the AFCS is it's brains. Hence you will physically feel your flight controls (cyclic/pedals & collective for 4 axis) move without your input or against your pressure as response to an input commanded by the AFCS/APLT. Similar to the SAS, you will see the tip path plane dance in response to the APLT input (long term stability) BUT you will also feel your flight controls move as well - the major difference to SAS input.
The APLT control authority is up to 100%. Because you feel the cyclic/pedals/collective move if the APLT makes an input to the disc, you can be made aware of these inputs and try to counter them by pushing against the force trim or pickle off the APLT (remove the brains) or Force Trim (remove its muscle).
Force Trim - is simply a means of holding the flight control position steady to allow the pilot to remove his hands from the flight controls in order to do other things in the cockpit without fear of the cyclic/pedals literally falling over and rapidly having the aircraft go out of control....ie the the UH1 without force trim on. It has up to 100% control authority if you were to continue to beep the coolie hat to change the disc/attitude, it will continue to input a stick change to displace the disc attitude. It is used as the "muscle" by the APLT function of the AFCS.
SAS actuators - Because the SAS requires a limit to it's control authority (eg 5-15% is typical), it has its own set of SAS actuators which have limited control authority, separate from the Force Trim (which has 100% control authority). This system provides NO feedback to the cyclic/pedals/collective if it generates a change to the disc attitude.
So I would start from scratch and dispel a few myths here. For one the autopilot does not "fly" the aircraft through the trim motors. I will explain.
So take a raw aircraft and let's give it a role - a private helicopter to take your gran up for a flight. What do you want from it? You want to be able to land in her garden, take off, flight around smoothly so she doesn't get ill then come back safely to the field. She natters a bit which is distracting so you want an aircraft that stays pretty much the same way up and pointing the same way as when you left it. You want nice predictable responses too.
So we basically want a really stable helicopter. When it's disturbed by an external force, you want it to resist and not be blown about. You want stability. Fine, get one of those contra-rotating rotor toys which pretty much stay still.
But this is no good as we want to move around. Therefore we need control. We want to be able to tilt the rotor to tilt the lift vector, we want to control heading via the tail rotor and control up and down/power with the collective.
But we don't want too much control. Consider the R22. If you take your eye of the ball for 2 secs off it wanders. But when you put an input in, boy does it move. But it takes a definite amount of cognitive effort to process the visual cues, (attitude change), think what needs to be done, command your limbs to move, make the movement and then process the cues again (the pilots control loop). Due to human limitations, this long loop can get out of synch with the aircraft giving a PIO.
SAS (stability augmentation system) is designed to help you. Without a SAS, the response of many aircraft to intentional or unintentional changes is an accelerating response (the rate of change of attitude gets bigger over time) towards a relatively large final attitude state. This is hard for your brain to process (although with a lot of attention (R22 hovering) you can). What SAS does through some type of rate sensing gyro (which could be a solid state AhRS in the modern world) is to sense changes in attitude rate and damp them down. Thus with a gust of wind, the rate is damping such that actually the aircraft doesn't react much. Of course the vertical fin does the same thing aerodynamically (in some case a SAS is sed to compensate for an inadequately sized fin - eg the T3/P3 variant of the EC135). But of course what if you make an intentional input and the SAS damps that out: too much stability and not enough control!
So SAS has to recognise when you make an input too and damp this appropriately. What you feel is that the response is lower in magnitude but it gets to this steady state faster. You would describe this as the controls feeling crisp.
SAS actuators do operate on an inner loop to achieve this. They have their own internal sensors which sense the parameter (usually rate of change of attitude), formulate a response, drive the actuator and then sense the parameter. There's your inner (fast) loop. You want the actuator to response quickly (so it does not get into a PIO like you) but what if this ran away by accident? That is why SAS actuators are restricted to 5-15% of travel. However this means they can get saturated (reach full travel). Generally SAS will have some indication for the pilot so he knows this has happened and he can move the appropriate control...this helps put the SAS actuators back in the centre of their travel and let them work again. Some pilots won't like SAS as it has a damped feeling and they cannot achieve the rates their used to but they have to work harder.
I will follow on with autopilot in next post.