KAP-140 questions

[gmainwaring]

I just bought the A2A C172 and am generally very impressed. I've been disappointed with other simulators' lack of realism. This C172 feels a lot more like the real airplane than anything I've seen in a non-certified personal simulator. Great job!

I'm a bit confused on some of the details of the included KAP-140. It behaves somewhat differently than the one in my airplane, and I'd like to understand why.

First of all, when the A2A KAP-140 is engaged in any mode, the control yoke ceases to have any effect. In my real airplane, this is not the case - you can override the autopilot with yoke inputs, and if you do so for more than two seconds, the autopilot will disengage. This is not recommended, and in the normal course of events you wouldn't ever want to fight against the servos like this. But if you set a climb, this autopilot will blindly fly you into a stall and kill you ... so it's pretty important that when you shove forward, you're able to overcome it. Yes, there's an AP Disco button on the yoke, but in the moment you might not think of it. I believe this is true for elevator control even on the KAP-140 with control wheel steering ... but the A2A KAP-140 clearly doesn't have this, because turning the yoke has no effect, and with CWS it would at least move the autopilot heading.

Second, the interaction between the A2A KAP-140 and the RealityXP GTN 650 is considerably different than I would expect. It works exactly as it should when not using the GPS, for example intercepting a VOR radial with the CDI selector in VLOC mode. You establish an intercept heading, switch to ROL mode, and press NAV; the unit flashes HDG to remind you to turn the heading bug to the inbound course, then sequences to NAV ARM. Once you arrive at the radial, the autopilot makes the turn inbound and to the heading from the heading bug, enters NAV mode, and keeps the CDI centered (which, if there is wind, will drive it away from the heading bug heading). As expected, entering NAV from HDG mode is the same except it selects a 45-degree intercept, and APR mode is the same except it is more precise and will intercept a glideslope via the GS ARM - GS sequence. All this is modeled perfectly.

In my airplane, none of this changes when the CDI is in GPS mode. You still have to set up an intercept, sequence through NAV ARM, etc. Communication between the GPS and the autopilot is entirely through the deflection of the CDI needle - the system cannot anticipate a turn, and just follows the needle and the heading bug, as it would with a VOR. The A2A KAP-140, on the other hand, behaves quite differently when the CDI selector is in GPS mode. When you press NAV or APR, there is no sequence through an "ARM" mode - instead, the system just goes directly into NAV or APR mode, and steers to the the GPS DTK regardless of the heading bug setting. Basically, it acts like a GFC-700 or GPSS-enabled S-Tec 55.

Although mine doesn't, some KAP-140 equipped 172s do have a form of GPSS capability. The way this works is that you install a translator box between the GPS and autopilot. This gives you a button or switch on the panel that selects the source for the autopilot heading input - it can either come from the heading bug or the GPS DTK. With this switch in GPS mode, you put the KAP-140 into HDG mode, and it follows the GPS - essentially, the GPS is turning an imaginary heading bug to tell the autopilot what to do. But even with such a system, the NAV and APR modes of the KAP-140 still work as usual, with the flashing HDG, sequence to ARM before going to NAV/APR mode, etc. If you want it to intercept a glideslope, you still have to baby it along into APR ARM and then APR mode.

I'm curious whether the A2A KAP-140's GPS implementation is the way it is because it's simulating some particular real-world implementation I'm not aware of, and if so, what that is - or if it's non-realistic due to some inherent limitation of the simulator, or just because it isn't modeled correctly. (Which would seem odd, considering how meticulously everything else is modeled.) In case it matters, I'm running this with Prepar3D 4.5.13 and RealityXP GTN 650 2.5.22.0.

Thanks in advance for any information you can provide.

[Nick - A2A]

Welcome to the A2A forums gmainwaring and thanks for the post.

The KAP 140 simulation in the Accu-Sim 172 has been coded independently to the default FSX/P3D autopilot, so for the most part, it should act just like the real thing.

However, as you've noted, there are one-or-two exceptions to this. A KAP 140 installation with CWS isn't modelled, and the ability to overpower the servos with yoke inputs also isn't simulated. In the P3D environment, it's worth bearing in mind that the relationship between the position of our physical controllers and the control surfaces themselves is a bit different to real life. For example, when we apply trim in the simulator, our physical yoke or joystick isn't deflected as it would be in the real aircraft. If every flight simmer owned force-feedback hardware, it would probably be feasible to model this behaviour a bit more realistically, but for now, we've had to make a few compromises in this area.

Add-on GPS units are another area where some compromises may be necessary due to the wide range of simulated and real life options. We try to support a variety of devices from third-party publishers, most of which were actually released to the market after the Accu-Sim 172 and its KAP 140 simulation were developed. The various products from different publishers behave slightly differently, and are continually being tweaked and updated by their developers just to add to the challenge of making sure everything behaves as realistically as possible.

You may find the discussion here of interest too. Seems the OP in this case was making the point that GPSS should indeed be implemented based on his particular 172. His sim installation would have used the Flight1 GTN, as the RXP version wasn't available at the time the thread was posted.

Thanks,
Nick