I'm trying to understand the differences here between an autopilot that is "coupled" into a GPS/NavCom like a Garmin 530W and what GPSS and altitude hold (like in an S-TEC 50 autopilot) have to do with this. Can anyone explain this to me? Maybe there is no difference, I really need this explained.
Pugz provided some very good information back on September 29th. As I recently installed a Garmin GTN 650 and add S-TEC GPSS to my 35 year old S-TEC 60-2 autopilot I thought I might respond also. As I did this installation myself I am quite familiar with the wiring involved which helps in understanding how the autopilot responds so differently to GPSS (GPS Roll Steering) than it does to the CDI (Course Deviation Indicator). First we need to understand that after installing GPSS the autopilot has two completely different interfaces. When in NAV mode the autopilot responds to the CDI for lateral guidance (and the CDI glide slope if the autopilot has that capability like my 60-2 does.) When in HDG mode the autopilot would normally respond to the Heading Bug on the Directional Gyro but if GPSS is installed the autopilot actually has two separate functions while in HDG mode: 1) Follow the HDG bug on the DG; 2) Follow the GPS Roll Steering commands that are provided by the GPS via ARINC 429 output that is routed to the autopilot GPSS module and converted into Heading Commands as if the HDG bug were providing them. In order to implement this the GPSS module has a switch to change from HDG mode to GPSS mode. As far as the autopilot is concerned he is just getting heading commands from a HDG bug on the directional gyro. The autopilot has no idea that he is being driven buy the GPS while in GPSS mode. Just imagine that a very skilled pilot is operating the HGD bug on the DG and the pilot knows exactly when and how much heading change to make in order to turn smoothly on the next route segment or intercept that route segment if off course. This is exactly what is happening except that the very skilled pilot happens to be the GPS receiver. The GPS knows exactly what heading changes need to be made and when to make them in order to guide the aircraft around a procedure turn and intercept the final approach course.
The new Garmin GI-106B CDI that I installed along with the GTN 650 still has some very important functions. After the GPS has guided the autopilot (via GPSS) through the entire route and possibly a procedure turn course reversal and intercepted the final approach course I switch the autopilot from HDG mode to NAV mode in order to engage the autopilot on the CDI. Now if I am flying an ILS or RNAV LPV approach the autopilot has both vertical and lateral guidance via the CDI. The CDI will either be getting input from the ILS signal or from the GPS depending on what has has been selected on the GTN 650. Both autopilot interfaces are still important (HDG/GPSS & CDI) but the autopilot CDI interface is used only for the final approach where vertical guidance is needed. Most of flight is best flown with the autopilot engaged in HDG mode either flying on the actual heading bug (being vectored) or with GPSS engaged and the GPS is turning the virtual heading bug for you in order to fly the programmed route.
An autopilot in NAV mode, using GPS as a source (GN*430+), will do the same thing as an autopilot using GPSS directly from a GPS (for example, KLN89/90 or GPS155XL).
GPSS is, as I understand it, a holdover from the time when panel-mount IFR-approved GPS units were standalone. GPSS was the way to feed your autopilot course correction data from that dedicated device. When the Garmin GNS430 hit the market and utterly destroyed the competition, GPS/NAV/COM became the standard, and the GPS or CDI control determined what the data was going to the autopilot.
In some aircraft you get into odd situations where a newer panel is paired with an older autopilot. I have a fair bit of time in an AG-5B Tiger with G1000, and it has the S-Tec System 30 autopilot with a GPSS module. I never understood this setup, as the System 30 has low/high VOR/LOC/GPS tracking built in. The GPSS module was completely redundant and seemed to be a holdover from the pre-G1000 model of the aircraft.
Source: personal experience, made more certain via this thread on the Avidyne forums
GPS units paired with older autopilots such as the S-TEC 30/50 mentioned here do not automatically give GPSS steering to the autopilot unless the GPS has a GPSS system built in. In this case the additional GPSS unit swaps between traditional NAV/APR/BC modes that seek using deflection from the CDI and GPSS steering which is data-driven straight from the GPS. It is possible to get a PFD that has GPSS built in as well (Aspen Evolution) and takes care of this functionality.
A functional difference between these two systems is demonstrated as such with a Garmin GNS-530W/430W and an STEC-50 autopilot with the separate GPSS unit:
1) with the GPS giving guidance and GPSS off (or not installed) the autopilot treats the GPS input signal as a CDI input - in other words, if there is a course change of greater than 10 degrees in either direction the autopilot will follow no guidance and do whatever it wants to, for example ignoring a 90 degree right turn in the flight plan (other than making an initial heading change of a few degrees while attempting to track as the CDI quickly goes full deflection) and then continuing on in whatever random way it wants without following guidance or a flag showing on the CDI (there is no flag because the guidance is present, but it's out of range just like a VOR signal would be!). Accordingly all major course changes must be executed manually or in heading mode by hand then the autopilot is engaged in NAV mode to track the course once it is captured and within 10 degrees of the present course.
2) with the GPS giving guidance and GPSS on the GPSS unit sends direct roll steering commands to the autopilot that are extremely accurate. The autopilot must always be in HDG mode to accept these inputs. The heading bug on the DG or HSI is effectively ignored while GPSS is engaged. Here's an example: while tracking a course (or really at any point in flight), selection of a 90 degree right turn (say, direct to) in the GPS will result in direct roll steering commands being given to the autopilot and the autopilot executing those commands and following them extremely closely within the parameters of the autopilot design (standard rate turn, etc.), resulting in an automatic turn to the new course, including automatic intercept of that course if the GPS performs that function, and any following courses and changes being flown precisely and automatically as they are sequenced in the flight plan.
Source: I now own an aircraft that has this system and operate it all the time.