Interesting EFIS Observation

[Ironflight]

Under the heading of ?knowing how your stuff works?, here?s a bit of trivia on the Dynon ?D?-series, Tru Trak ADI Pilot, and GRT Horizons. We recently had the pitot/static/xpnder/IFR certs done on both of our airplanes. For those who haven?t done this, it involves plugging into the pitot/static systems and using a vacuum pump and a calibrated test set to take the instrumentation system up to a simulated altitude of 20,000? (or greater, in some cases). This checks the static system for leaks, checks the altimeter for calibration, and makes sure that the transponder is reporting codes and altitudes properly. In our case, everything passed with flying colors once again (whew!), but I was able to make an interesting observation that tells me something about the way Attitude is calculated in the systems.

With the test set-up hooked to the RV-6 (Dynon D-10A & D-180, Tru Trak ADI), as we started ?up?, all three attitude indicators showed a significant pitch up ? about 15 degrees nose high with a 2,000 fpm climb rate. This was, of course, with the airplane sitting still, and not changing attitude. Clearly, this is not an expected flight condition, but it does tell me that pitot/static does play a part in the attitude solution for these instruments. A pitch down was indicated during ?descent?, but I don?t think it was quite as pronounced. When we transferred the test rig to the RV-8, with GRT equipment, there was no indication of a pitch change at all, even with a 3,000 fpm climb and descent. So apparently, the attitude algorithms that GRT uses don?t care about pitot/static. I have not seen the attitude determination software code for any of these systems, but I know this has been talked about before on the forums ? how the various electronic attitude systems are ?aided? to get their solution. So this isn?t really new, just an observation that you can see it in action with this kind of a test. Interesting from an engineering standpoint, and good to know that a static leak might affect the attitude solution with some equipment (although I doubt that in real life, it would be a huge error).

Paul

[flickroll]

I was present (assisted the Tech actually) during the IFR cert on a dual AFS system installed in my RV-8 (yes, a dual cert costs more ). The AFS system is like the GRT in that it did not indicate a pitch up or pitch down as Paul said happened with the Dynon system.

[dynonsupport]

If connected to a GPS, the Dynon EFIS systems can use that as a backup to airspeed. So in flight, you would not see this behavior with a failure of the airspeed system. For engineering reasons, this is not active until we see airspeed on the GPS at least once, which is why it was not active during your test. If you did this test after a fast taxi or a full flight you would not have seen any motion.

As for the TruTrak ADI, all owners should read up on how it works, since it is very important to know that it is not an attitude device, it's a VSI with some very short term pitch rate biasing.

[Brantel]

Of course they all have different algorithms driving their attitude display, different ways to aid/correct the affordable gyro's used in these systems, I am sure there are pros and cons of doing each way and in this case, who is to say which way is better?

I do know that some ways are better than others in regards to how fast a AHRS will recover from being restarted, saturated and tumbling and being able to recover in flight...It is my understanding that some take a long time to recover and some may never recover while in flight.

The Dynon can and will recover in flight in a small amount of time.

Like Dynon sez, the TT ADI is not a true attitude indicator and it works off of vertical speed in pitch so if you fake it into thinking it is going up or down, it will indicate pitch up or down. That being said, in most flying conditions it still works good enough to be a good backup for attitude.

[Jamie]

Originally Posted by dynonsupport

If connected to a GPS, the Dynon EFIS systems can use that as a backup to airspeed. So in flight, you would not see this behavior with a failure of the airspeed system. For engineering reasons, this is not active until we see airspeed on the GPS at least once, which is why it was not active during your test. If you did this test after a fast taxi or a full flight you would not have seen any motion.

As for the TruTrak ADI, all owners should read up on how it works, since it is very important to know that it is not an attitude device, it's a VSI with some very short term pitch rate biasing.

I plumbed in a small valve into my pitot system to test the GPS aiding back when it was first introduced into the firmware. The GPS aiding works well, with one (very big) caveat.

The GPS aiding will only work IF your pitot pressure drops to static pressure (meaning that your IAS goes to 0). That is to say, for example if you are in the clouds and your pitot ices over and your pitot drain is open the GPS aiding will work well. If the drain is clogged, GPS aiding will NOT turn on.

Dropping the IAS to zero was the only way I could make the GPS aiding work. A partially clogged pitot or a significant pitot leak (tested with my valve temporarily plumbed in) resulted in pretty dramatic pitch deviations of the EFIS and no indication of GPS aiding.

As a result I added another small valve permanently to my pitot system (sort of like alternate static). If my pitot ever becomes partially clogged or leaky I can open up the pitot line and GPS aiding will kick in.

The simplest fix would be to add a menu option to manually switch to GPS aiding mode.

[Alan Carroll]

Originally Posted by Brantel

Like Dynon sez, the TT ADI is not a true attitude indicator and it works off of vertical speed in pitch so if you fake it into thinking it is going up or down, it will indicate pitch up or down.

Agreed, but this could also be construed as better than a true attitude indicator in some ways? Trutrak's underlying philosophy seems to be that pitch attitude and magnetic heading are not as important as your actual path through the air (vertical speed and track). Most instrument systems require that the pilot interpret how the former is affecting the latter, and this hasn't fundamentally changed since the late 1920s. The ADI gives you flight path directly, which in principle should make piloting easier. This is not how we were trained however...

In several hundreds flight hours I have yet to see the ADI indicate an incorrect (up vs. down) attitude. It would require very low airspeed to fake it out, at which point a warning is flashed. A leak in the static system might cause problems though?

[Jamie]

It could also be argued that the ADI is a better flight instrument than an attitude indicator. When flying instruments with a traditional attitude indicator (or even an EFIS attitude indicator) the attitude indicator is not the primary pitch instrument in straight and level flight. Why? Because it changes based on the loading and speed of the airplane. The ADI, however, indicates a level attitude the same way, regardless of speed or AOA.

[dynonsupport]

Jamie,
It is true that GPS aiding will not turn on until airspeed goes to zero. We did this because we found very little error in our testing when airspeed errors were high. So if you are going 160KTS and your ASI is 75 KTS, the Dynon will still be quite accurate and will never diverge (show you level when you are not). Because of this testing, we only use GPS aiding when the pathological case occurs and all airspeed info is lost.

[Jamie]

Originally Posted by dynonsupport

Jamie,
It is true that GPS aiding will not turn on until airspeed goes to zero. We did this because we found very little error in our testing when airspeed errors were high. So if you are going 160KTS and your ASI is 75 KTS, the Dynon will still be quite accurate and will never diverge (show you level when you are not). Because of this testing, we only use GPS aiding when the pathological case occurs and all airspeed info is lost.

Interesting. That's definitely at odds with my findings. When straight and level, yes, it would stay there but I found that turning 45 degrees it never would recover properly until IAS dropped to zero or I closed off the 'leak' in the pitot system.

It was a while back when I did this, but if I remember correctly I tested it at around 100kts IAS while I was doing 167kts TAS (my typical cruise speed).

[penguin]

Originally Posted by Ironflight

Under the heading of ?knowing how your stuff works?, ... Interesting from an engineering standpoint, and good to know that a static leak might affect the attitude solution with some equipment (although I doubt that in real life, it would be a huge error).

Paul, excellent write up.

I think this phenomenon shows potentially a more significant problem as the attitude platform appears not to be very stable under any kind of failure condition. Clearly Dynon provides great equipment for VFR operations, but for regular IFR? To achieve their price point perhaps Dynon have had to make savings somewhere, perhaps this is one area? In my view the attitude solution should not require pitot or GPS aiding to remain stable.

Pete