Garmin have announced this recently:
I have the S-TEC55X autopilot, 2*GNS430W, G500 display. As I understand it this autopilot derives some inputs from the Turn Coordinator as opposed to an encoding AI. If I installed a G5 then presumably if I installed the GAD29B Adaptor as well then the 55X would receive the Turn Coordinator data from the G5 instead? Also, the standby vacuum driven AI I have could be removed as the G5 would take over the that function too meaning I could remove the vacuum pump completely? Have I got this right because if it’s correct it’s a no brainer and I’ll do it.
This is a mutch made mistake.
The G5 with GAD29B only provides HSI interface to the AP (HDG bug error and course pointer error output)
The G5 (and GAD29B) are not able to provide attitude nor turn rate information for third-party autopilots
(also no Alititude , airspeed, baro, magnetic heading output)
According to the installation manual, the 55X is compatible but is the only S-TEC autopilot for which GPS steering is not supported.
It did sound to me like the G5 + GAD29 could replace a KI256 (for e.g. a KFC225) which would have been a killer feature.
Technically this is easy enough to do, using any certified or uncertified AHRS which has some kind of pitch and roll output, and emulating the two LDVT coils of the KI256, but nobody dares trying to get the STC for it. Well, there is the G500+GAD43, EFD1000+EA100, G600. That’s about it I think. All of these are unattractive to many people for one reason or another. Replacing a KI256 with a G5 would be a great piece of isolated surgery.
What Garmin seem to have done is got an STC for the G5 to emulate an HSI which is, wait for it, a voltage representing the heading bug deviation from “straight ahead” and the course pointer position which is usually X/Y/400Hz. This is 1960s/1970s stuff. Maybe there is more? In certification terms this is probably massive 
STOLman wrote:
Pilots can simply select GPSS on the G5 and heading mode on the autopilot and the autopilot will fly smooth intercepts, holding patterns, procedure turns and more.
Noob question : doesn’t GPSS require turn coordinator information ?
The gyro driving my KCS55a HSI looks to be on the way out, so I’m going to update to the G5 ASAP. I was (cost permitting) going to replace the KI256 AI. It seems due to the flight director this isn’t possible. Garmin suggest ditching the turn co-ordinator and putting the KI256 in its place which is what I might do.
It’s a bit misleading… you have to read it carefully. The biggest “fish to fry” (KI256 emulation) is not delivered.
… Just realised this was posted earlier in this merged G5 thread
What is preventing Garmin from emulating the KI256? Garmin has a) the technical knowhow and experience to interface with 3rd party APs, and b) many years of experience working closely together with certification authorities, c) the financial resources to push projects through certification, d) a likely market. Is Garmin afraid of cannibalization?
Certification requirements? Remember that the G5 STC is quite limited as it comes from the experimental division.
Different strategy – promote the GFC500 ?
The G5 doesn’t emulate pitch and roll outputs to ANY autopilots other than Garmin’s own GFC500, so it’s not just the KI256 that’s an issue.
