Aspen AOA

[PT20J]

I'm curious about other's experience with the Aspen AOA indicator. Mine, in a 1994 M20J, was installed for the previous owner. It has never indicated properly -- a normal approach ends up with the pointers near or in the cross hatch area where they are supposed to be when the stall warning sounds. I checked the configuration info and re-flew the calibration procedure yesterday and it's still about the same. I sent all the data to Aspen tech support which sent it to engineering and I'm waiting for a response.  

In the meantime, has anyone got one to work right, and if so did you need to add bias values or tweak anything?

Skip

[mike_elliott]

I calibrated one once on the Bravo I had, it worked as advertised, Skip. It sounds like something is off on the data entered prior to calibration, but obviously cannot say for sure. I am sure Engineering from Aspen will be able to resolve.

[Marauder]

1 hour ago, PT20J said:

I'm curious about other's experience with the Aspen AOA indicator. Mine, in a 1994 M20J, was installed for the previous owner. It has never indicated properly -- a normal approach ends up with the pointers near or in the cross hatch area where they are supposed to be when the stall warning sounds. I checked the configuration info and re-flew the calibration procedure yesterday and it's still about the same. I sent all the data to Aspen tech support which sent it to engineering and I'm waiting for a response.  

In the meantime, has anyone got one to work right, and if so did you need to add bias values or tweak anything?

Skip

Skip - there is a step in the calibration section that talks about obtaining calibration values from Aspen. Let me see if I can find the document and post the section. 

[Marauder]

Skip - I suspect that some if not all of the pre-calibration values may be entered incorrectly or you need a Bias value from Aspen.

Check out the AoA installation manual. Look at pages 382 and 383 for the pre-calibration entries required.


https://aspenavionics.com/documents/webrepost/ADSB%20docs%20FF/900-00003-001%20BH%20EFD1k-5c%20SW2%20X%20Instl%20Manual.pdf


Sent from my iPad using Tapatalk Pro

[PT20J]

I checked all the calibration info per the manual and it was correct. I only changed the weights because they were different for my calibration flight than the calibration after installation. There are no mods on my aircraft and it stalls at the book speeds and I checked that the stall warning is calibrated within spec. The software should be up to date as it was checked last January when the ADS-B unlock was installed.

I sent Aspen tech support my test data and asked about bias values. The reply was:  The “In-Flight Pre AOA Calibration Check” and bias numbers are only used for aircraft with mods that are not accounted for in the AFM, like some types of VGs and STOL kits or with conditions specified in the Limitations section of the instructions. I’ll send this over to engineering and see what they say about using the bias numbers. Maybe they can pickup on what is causing the pointer to read so high.

Skip

[Marauder]

I checked all the calibration info per the manual and it was correct. I only changed the weights because they were different for my calibration flight than the calibration after installation. There are no mods on my aircraft and it stalls at the book speeds and I checked that the stall warning is calibrated within spec. The software should be up to date as it was checked last January when the ADS-B unlock was installed.
I sent Aspen tech support my test data and asked about bias values. The reply was:  The “In-Flight Pre AOA Calibration Check” and bias numbers are only used for aircraft with mods that are not accounted for in the AFM, like some types of VGs and STOL kits or with conditions specified in the Limitations section of the instructions. I’ll send this over to engineering and see what they say about using the bias numbers. Maybe they can pickup on what is causing the pointer to read so high.
Skip

Something isn’t right. Since it is purely a calculated AoA there has to be something set incorrectly. Hopefully they can get to the root cause.


Sent from my iPad using Tapatalk Pro

[Blue on Top]

Stalking.

[PT20J]

11 hours ago, Blue on Top said:

Stalking.

Well, since you are lurking about....

I took a look at the Aspen patent (attached) and thought over the calibration procedure. The calibration requires straight and level flight at two airspeeds (cruise and approach) clean, and at approach airspeed with gear and flaps down. Tolerances are pretty tight: airspeed constant to within 3 KIAS. vertical speed no more than 100 fpm, pitch change no more than one degree, roll change of no more than two degrees. (Side note: in smooth air, I was amazed at how well my 45-year-old design KAP 150 autopilot did that).

The object of the clean tests appears to be to determine the lift curve and the third test is to determine the lift curve offset with gear and flaps down (the lift curve slope would be assumed to be the same in both configurations). One of the primary objectives of the algorithm appears to be to avoid the necessity for supplying the aircraft weight during normal operation. I can see why this would be beneficial as having to enter the weight would be error prone and make the system more complex to use than competing systems.

The algorithm gets the gross weight stall speeds from the airspeed data entered in IAS CONFIG B. In this case, Vs is 62 KIAS and Vso is 58 KIAS from the POH. Assuming that the stall warning speed is 5 kts higher (which measurement shows to be the case), this gives stall warning speeds of 67 KIAS clean and 63 KIAS dirty at 2900 lb. Adjusting for the flight weight of 2321 lbs decreases these speeds by a factor of SQRT(2321/2900) = 0.895 or 59.9 KIAS clean and 56.4 KIAS dirty. However, during the pre-calibration flight test, I measured stall warning speeds of 56 KIAS clean and 52 KIAS dirty which are about 4 kts slower than the calculated speeds. (Aspen has you measure the stall warning speeds rather than the actual stall speed. This makes sense to me since the lift curve is highly non-linear around stall. It does introduce a source of error if the stall warning is not set correctly, but I checked mine and it is pretty close to 5 knots above stall in all configurations).

It seems pretty clear that the purpose of the bias values is to correct for non-standard stall speeds. Mine are only off about 4 kts, but maybe that's enough to mess it up. I'll wait and see what Aspen comes back with.

Skip

WO2016164624A1.pdf

 

[Blue on Top]

@PT20J  This is EXTREMELY interesting ... and a lot of information.  I need to keep my engineering hat off and my mouth shut when I look at this stuff ... lest I say something that will offend someone.  But when has that stopped me before?  

I love Wolfgang Langewiesche's "Stick and Rudder" but AOA is not the angle made by the "wing chord" and the direction of flight.   The relative wind is not equal and opposite to the direction of flight .... except in a computer.  Leighton Collin's "Takeoffs and Landings" is a good book.  To put this in perspective (no, not Garmin in a Cirrus), the forward to Leighton Collin's book was written by none other than Wolfgang Langewiesche.  They were really good friends.  Leighton is Bob Collin's father and was the originator and publisher of "Air Facts" magazine.

There is a good reason that certificated (and military) AOA systems are of two/three types ... they work ... and are designed and proven to be calibrated to the very high accuracy required.  They are also expensive.

Perception is EVERYTHNIG in sales.

Continuing to stalk.  I have to know what people are saying  

[LANCECASPER]

8 hours ago, Blue on Top said:

Leighton is Bob Collin's father and was the originator and publisher of "Air Facts" magazine.

Did you mean Richard Collins instead of Bob?  https://www.flyingmag.com/former-flying-editor-in-chief-richard-collins-dies/

[Blue on Top]

1 hour ago, LANCECASPER said:

Did you mean Richard Collins instead of Bob?  https://www.flyingmag.com/former-flying-editor-in-chief-richard-collins-dies/

OMG, yes!!!  I knew that something  was wrong when I typed it.  Thanks for the catch.