Jump to content

Recommended Posts

     Update on some knowledge gained after speaking with an engineer from Uavionix. I’ve got two of the AV30C’s in my Ranger and have had the DG precession issues as well as the AI acting like a vacuum system on its way out. 
     The firmware update has been released for the Experimental version and is awaiting approval by the, surprise surprise, FAA. It will involve either wiring into the RS232 interface with a connected laptop or the more ideal way of removing the individual units and installing the AVLink in each one. The -Link is essentially a tiny WiFi receiver and once installed will only need a computer nearby to wirelessly transfer firmware files. The engineer said the Link would be inexpensive and future proof so any future capabilities certified could be easily added.

     The current update awaiting approval MAY fix the precession issue for Mooney’s. Apparently our aircraft are giving them trouble that other types are not. There is some question as to whether a Mooney with magnetized tubing might be having some affect. Merely the messenger here, I’m not offering any data based opinion. It’s the same with my AI issue. Currently my shop swapped out the number one with another new AV30 and is awaiting my return to test fly, I’ll let folks know the results in another week or so.

     Another tidbit, Garmin devices continue to not play well with other brands. Apparently there is a proprietary signal from the 175/355 units that confuses the AV30 data fields for GPS destination. Mine show correct nav info like distance to go and cross track but will rarely display the actual waypoint identifier name. It displays “ALT” instead. Kind of annoying but another issue Uavionix is aware of and trying to fix.

     SO, jury is still out on the final performance of the certified AV30’s. I still like the company a lot, they have always been responsive in speaking with me even though it may not be the answers I desire. The engineer also admitted they have been facing headwinds with the FAA because they are a small company with tight pockets and aren’t able to muster the attention the big G can afford but they stay persistent. Traffic on the AV’s and autopilot interfaces are the next closest projects.

     Hope this helps some folks. I’m keeping mine for now,  no guarantees on what I do in the future. 

  • Like 2
Link to post
Share on other sites

*Members that donate $10 or more do not see advertisements*

I wonder if there are any software updates for the original AV20s units?

They did not seem to calculate true airspeed correctly and the angle of attack never worked on most Mooney’s. 

  • Thanks 1
Link to post
Share on other sites

Thanks for the Pirep on this, @Jakes Simmons.

The hypothesis about the magnetized tubing is interesting - I'm curious how much of an effect that would play.

I haven't installed an AV-30 myself, but I've heard plenty of reports from operations of just about every type (Cessna, Piper, Mooney, Beech, Vans, etc.) about the precession issue. My local FBO has them installed in their 172s and they've been battling the same issue themselves. None of the AV-30 manuals seem to reference any detailed operational theory so it really bugs me not knowing exactly what the source of this issue is.

If I may indulge in a bit of speculation, though, given the fact that uAvionix is working on a software approach to remedy this issue, I sometimes wonder if the AV-30 DG simply operates by using acceleration data pulled by an accelerometer (not a gyro) that is effectively double-integrated (∫∫) to obtain position (i.e. heading)? I've noticed that the data refresh rate of the AV-30 is painfully slow (I estimate something like 15Hz compared to the G5's 30+Hz) which leads me to wonder if this slowness is what is causing minute errors in calculation that add up over time? Of course, this is a hail-mary guess, and I could be absolutely off the mark for all that matters.

The Link interface seems pretty cool. We've definitely seen a leap in firmware-update technology over the past few decades in the world of GA avionics; whereas the firmware would need to be loaded via custom cable harnesses in the early 2000s, avionics started incorporating USB in the 2010s, and now we're moving towards ad-hoc Wi-Fi loading. Neat.

I'm not surprised one bit about Garmin though. They make excellent equipment, but starting in the early 2010s, they've become increasingly notorious for not playing well with outside data protocols, and more often than not, intentionally obfuscating and/or modifying data protocol details to prevent 3rd party compatibility. The ADS-B+ protocol is likely the most notable example, and the only reason other companies like Avidyne, Trig and Appareo are "compatible" with Garmin GPS sources is because these companies went to great lengths to reverse engineer the protocol. They've also changed the verbiage of their latest STC permission letters to preclude non-Garmin dealer technicians from installing used equipment in the future (even though these products would have fallen out of their Dealer Install Policy by that time), which points to a trend of incorporating planned obsolescence into their business practices.

Certification has been a particular painful spot for smaller manufacturers lately, especially those that can't afford in-house approval. ACO's across the country had largely been preoccupied with the 737MAX fiasco, and I suspect the ongoing 787 issue and 777/PW4000 incident won't help either.

Link to post
Share on other sites
19 hours ago, Minivation said:

is effectively double-integrated (∫∫) to obtain position (i.e. heading)?

Either +5 extra credit for being the first on MS to show a double integral, or -5 for same.   Gotta refer this one to the Dean of Mooneyspace, @carusoam.

 

  • Like 2
  • Haha 5
Link to post
Share on other sites

Sandia Quatro which are also BK units had issues with some of their AI units that triggered an AD. I don’t think that had been resolved yet.  Are the AV30 AI issues similar? It is worth noting that the Garmin units require a GPS input. I wonder if this keeps their units from drifting off?  But I have noted even a few Garmin G5 users have also reported issues. 

Link to post
Share on other sites

How are folks dealing with these issues when flying IFR?  Has anybody here flown much IMC with this set up?  I suppose the drifting DG is a nuisance, but it’s not so easy to compensate for ADI issues?  Is there a limitation for VFR use only?  Does not seem to be a problem limited to Mooneys only.  I’ve seen this reported on multiple sites.  I’m really pulling for them and have had my eyes on this for a while, but it does not seem ready for IMC.

Link to post
Share on other sites

I have about 20 hours with mine but no hard IMC yet.  Mine replaced the AI and as such has worked well.  However, the AV-30-C AI can also display the DG across the top, and keeping it "accurate" with the whiskey compass is a game in itself.  Since I don't need it, I instead display the GPS heading in the same spot just because.

Due to insufficient research on my part prior to purchase (I simply relied on uAvionix's webpage for info), I was not fully aware of the precession issue and have quite a bit of egg on my face.  Like @Jakes Simmons I still like the company a lot and am eagerly awaiting their approval for autopilot integration before I purchase another one to replace my vacuum DG with heading bug.  However, this time I won't be as quick to pull the trigger.

The WiFi link sounds cool.  As it is, access to the back of the instrument and disconnecting the existing D-sub connector is required for any upgrades.  We all know how fun it is to access the rear of a top instrument in a Mooney panel, especially if a 201-stype windshield is installed.   WiFi upgrades sound simple enough for a cave man to do it.

Thanks for the updated info.

  • Like 1
Link to post
Share on other sites
On 2/26/2021 at 1:10 PM, Gary0747 said:

I wonder if there are any software updates for the original AV20s units?

They did not seem to calculate true airspeed correctly and the angle of attack never worked on most Mooney’s. 

I wonder the same thing.   I was in contact with Aerovonics before uAvionics bought them with a request for a software update.  The response from Aerovonics was to not expect the AOA to work on a Mooney.   That doesn't really bother me because I bought it for the timer, TAS, and mems AI functions.

There is obviously a bug in the TAS calculation code, because it doesn't calculate it correctly   I've e-mailed uAvionics three times.  First time I got a "I'll pass your request to engineering..."  with no reply.  The next two e-mails both received no replies.  Wish they would reply to legitimate customer questions - they should.

At our Mach numbers the constant of proportionality between IAS and TAS with a few decent assumptions is sqrt(ρ/ρo), the square root of air density ratio, with ρ=Ps/RT.  The device measures Ps & T, and R is approximately constant within less than 2% for air.  Yet, the AV20s gives erroneously high TAS indications, at least in my case, which I find supremely annoying.  IAS to TAS conversion is easy as the little vernier scale on some AS indicators shows.  Perhaps someone forgot to take the sqrt() of the air density ratio when they wrote the firmware.   I'd help debug it if they would share.   Wouldn't an open source programmable NORSEE device like the AV-20-S be cool? 

  • Like 1
Link to post
Share on other sites
33 minutes ago, 0TreeLemur said:

I wonder the same thing.   I was in contact with Aerovonics before uAvionics bought them with a request for a software update.  The response from Aerovonics was to not expect the AOA to work on a Mooney.   That doesn't really bother me because I bought it for the timer, TAS, and mems AI functions.

There is obviously a bug in the TAS calculation code, because it doesn't calculate it correctly   I've e-mailed uAvionics three times.  First time I got a "I'll pass your request to engineering..."  with no reply.  The next two e-mails both received no replies.  Wish they would reply to legitimate customer questions - they should.

At our Mach numbers the constant of proportionality between IAS and TAS with a few decent assumptions is sqrt(ρ/ρo), the square root of air density ratio, with ρ=Ps/RT.  The device measures Ps & T, and R is approximately constant within less than 2% for air.  Yet, the AV20s gives erroneously high TAS indications, at least in my case, which I find supremely annoying.  IAS to TAS conversion is easy as the little vernier scale on some AS indicators shows.  Perhaps someone forgot to take the sqrt() of the air density ratio when they wrote the firmware.   I'd help debug it if they would share.   Wouldn't an open source programmable NORSEE device like the AV-20-S be cool? 

I have one of the AV-20s installed in my E model in hopes that it could just be used as a clock and that in a pinch it could be used for its AOA features as well as emergency source of attitude but the more I have read about it the more I feel it will just be an expensive clock.

@0TreeLemur certified and open source are a word combination that aviation regulatory bodies have not learned yet.

  • Like 1
Link to post
Share on other sites
3 minutes ago, AerostarDriver said:

I have one of the AV-20s installed in my E model in hopes that it could just be used as a clock and that in a pinch it could be used for its AOA features as well as emergency source of attitude but the more I have read about it the more I feel it will just be an expensive clock.

@0TreeLemur certified and open source are a word combination that aviation regulatory bodies have not learned yet.

I hear you and I agree that certified avionics should not be open source.   I want the vendors skin fully in the game.

The AV-20-S is NORSEE.  If they open sourced the code we could fix their bug.  They would still have to approve it and compile.   I suppose they would have to flash the ROM too.   The one I purchased has no USB port.

Link to post
Share on other sites

I installed an AV-20 in a friend's M20C back in 2019 and ... it didn't turn out well. I felt really bad for him but after days of frantic searching I found that my Mooney install wasn't the only one having issues. Judging by their theory of operation (AOA = AHRS pitch angle - FPA [tan(VS/IAS)] things should have worked out OK, but using the aforementioned equation, I can only surmise that either the AHRS got messed up (despite my best efforts to "zero" the pitch out at level flight attitude) or the IAS is far off from CAS (I know dropping the gear messes with the static system considerably, but does it that much?)

Anyway, a few days after the install, I attended OSH '19 and had the opportunity to talk with Bill Shuert, who was Aerovonics CEO at the time. Nice gentleman, but when I asked him about (a) any ideas about the erroneous AoA, and (b) any possibility for improving the data refresh rate of the AV-30, the answer was basically (a) not my problem and (b) no. It was that exact moment I resolved myself that I would not be recommending an Aerovonics product to a customer unless they insisted on them.

3 hours ago, 0TreeLemur said:

If they open sourced the code

image.png.79cd09a2f8bc385185f79456ad94160d.png

I've long pondered about the plausibility of porting over "open-source" to the avionics world. On a technical level, both computer and avionics industries are somewhat similar in that they involve computer hardware/software (well, more like a network of them) to accomplish a purpose or task, but the key differentiator is the market size - billions of people for the larger computer industry vs. several hundred thousand at most for avionics, despite the fact that the entry barrier for the latter is much higher (i.e. certification) and the risk of getting litigated to oblivion is equally ridiculous too. To add to that, if open-source software were to exist for avionics, you can certainly bet XYZ Corporation would snatch up that code, pretty it up, add some proprietary bells & whistles, then go to town to sell it. That's basically what Garmin did with their slew of RS232 protocols in the late 2000s.

To that end, I say all this with a greatly exacerbated tone. I am a huge fan of open-source and wiki-based projects as nothing bothers me more than half-assed businesses that overcharge and underdeliver. Of course, in the interest of safety, "crowdsourced" and "proven & reliable" don't often fit in the same sentence (I mean, good luck with anything past DO-178 Level D!) but that doesn't stop me from wondering sometimes "what if we set up a nonprofit avionics manufacturer with the vision & purpose of bettering GA's safety and affordability". Of course it's a pie in the sky, but one can keep dreaming.

 

Edited by Minivation
Link to post
Share on other sites
59 minutes ago, Minivation said:

I installed an AV-20 in a friend's M20C back in 2019 and ... it didn't turn out well. I felt really bad for him but after days of frantic searching I found that my Mooney install wasn't the only one having issues. Judging by their theory of operation (AOA = AHRS pitch angle - FPA [tan(VS/IAS)] things should have worked out OK, but using the aforementioned equation, I can only surmise that either the AHRS got messed up (despite my best efforts to "zero" the pitch out at level flight attitude) or the IAS is far off from CAS (I know dropping the gear messes with the static system considerably, but does it that much?)

Anyway, a few days after the install, I attended OSH '19 and had the opportunity to talk with Bill Shuert, who was Aerovonics CEO at the time. Nice gentleman, but when I asked him about (a) any ideas about the erroneous AoA, and (b) any possibility for improving the data refresh rate of the AV-30, the answer was basically (a) not my problem and (b) no. It was that exact moment I resolved myself that I would not be recommending an Aerovonics product to a customer unless they insisted on them.

image.png.79cd09a2f8bc385185f79456ad94160d.png

I've long pondered about the plausibility of porting over "open-source" to the avionics world. On a technical level, both computer and avionics industries are somewhat similar in that they involve computer hardware/software (well, more like a network of them) to accomplish a purpose or task, but the key differentiator is the market size - billions of people for the larger computer industry vs. several hundred thousand at most for avionics, despite the fact that the entry barrier for the latter is much higher (i.e. certification) and the risk of getting litigated to oblivion is equally ridiculous too. To add to that, if open-source software were to exist for avionics, you can certainly bet XYZ Corporation would snatch up that code, pretty it up, add some proprietary bells & whistles, then go to town to sell it. That's basically what Garmin did with their slew of RS232 protocols in the late 2000s.

To that end, I say all this with a greatly exacerbated tone. I am a huge fan of open-source and wiki-based projects as nothing bothers me more than half-assed businesses that overcharge and underdeliver. Of course, in the interest of safety, "crowdsourced" and "proven & reliable" don't often fit in the same sentence (I mean, good luck with anything past DO-178 Level D!) but that doesn't stop me from wondering sometimes "what if we set up a nonprofit avionics manufacturer with the vision & purpose of bettering GA's safety and affordability". Of course it's a pie in the sky, but one can keep dreaming.

 

Might not be that far off....at least in the experimental world.  You are probably well aware of the Stratux ADS-B In device.  That lead to some success with crowd sourced non-certified avionics.  I continue to use one, but the AHRS version has not been as successful in the open source product.  I have to wonder if it is because some of the key collaborators have moved on to start other operations like this:  https://www.falkenavionics.com. (...the old FlightBox)  I certainly agree with your sentiment in the DO-178 world...

 

Link to post
Share on other sites
4 hours ago, Minivation said:

Anyway, a few days after the install, I attended OSH '19 and had the opportunity to talk with Bill Shuert, who was Aerovonics CEO at the time. Nice gentleman, but when I asked him about (a) any ideas about the erroneous AoA, and (b) any possibility for improving the data refresh rate of the AV-30, the answer was basically (a) not my problem and (b) no.

 

Yes, now that you mention the name, Bill Shuert was the person who advised me to de-activate the AoA in my AV-20-S.

Link to post
Share on other sites
  • 5 weeks later...

I've noticed that on the first flight of the day, my DG drifts a lot as mentioned by others. On the second flight, it is rock steady, and so I'm pretty confident that this firmware update will in fact fix the drift issue. The start/stop instructions given by the company didn't really help to mitigate the issue in the meantime, however.
Overall, I'm pretty happy with my setup (see picture). The AI is reliable. I regularly cross-check with my vacuum driven AI and they are always in agreement (in the picture, the vacuum driven AI looks nose-down due to parallax). I wouldn't hesitate to take it into IMC, although I will wait until the firmware upgrade is out. It has been a great upgrade on the cheap. It cost me about 4.5k to get both installed and while an HSI would be nice, the distance to waypoint, waypoint bug, track, desired track, etc give a lot of situational awareness without having to glance at the box.

A couple of things I would like to see improved. While there is a "selected altitude" option, there are no callouts sent to the audio panel when you get close or deviate. Kind of surprised they did that, the selected altitude is essentially just a reminder for you to include in your scan, along with the current altitude.
Another thing is that the AI scale is a bit coarse in my opinion. Having a finer scale would help a lot in IMC. Currently, the "aim small, miss small" approach to flying with an AI is very difficult, since the scale is just not sensitive enough to display small changes in pitch.

Overall, happy with my purchase.
20201205_131256.jpg?psid=1&width=697&height=929

Link to post
Share on other sites

I have the same setup in mine and the DG wonders as well. The information, as of this morning, was to let the instruments "warmup" prior to starting the engine. I'm in annual fighting a power issue so it will be a bit before I can test the advice.

I agree with the call outs previously made, but I would also add my appreciation for the vertical trend. Call me lazy, but I can dial in my trim a lot faster and have it locked in a fraction of the time. 

It would be nice if the wifi was available for these updates when they get released.

  • Thanks 1
Link to post
Share on other sites
5 hours ago, VinceCB said:

I've noticed that on the first flight of the day, my DG drifts a lot as mentioned by others. On the second flight, it is rock steady, and so I'm pretty confident that this firmware update will in fact fix the drift issue. The start/stop instructions given by the company didn't really help to mitigate the issue in the meantime, however.
Overall, I'm pretty happy with my setup (see picture). The AI is reliable. I regularly cross-check with my vacuum driven AI and they are always in agreement (in the picture, the vacuum driven AI looks nose-down due to parallax). I wouldn't hesitate to take it into IMC, although I will wait until the firmware upgrade is out. It has been a great upgrade on the cheap. It cost me about 4.5k to get both installed and while an HSI would be nice, the distance to waypoint, waypoint bug, track, desired track, etc give a lot of situational awareness without having to glance at the box.

A couple of things I would like to see improved. While there is a "selected altitude" option, there are no callouts sent to the audio panel when you get close or deviate. Kind of surprised they did that, the selected altitude is essentially just a reminder for you to include in your scan, along with the current altitude.
Another thing is that the AI scale is a bit coarse in my opinion. Having a finer scale would help a lot in IMC. Currently, the "aim small, miss small" approach to flying with an AI is very difficult, since the scale is just not sensitive enough to display small changes in pitch.

Overall, happy with my purchase.
20201205_131256.jpg?psid=1&width=697&height=929

The Garmin G5s have altitude select as well but don’t provide audio either.  I think if you get the full autopilot you might get the altitude selection audio but I’m not sure.

  • Like 1
Link to post
Share on other sites
On 4/1/2021 at 2:59 PM, Shiver said:

I have the same setup in mine and the DG wonders as well. The information, as of this morning, was to let the instruments "warmup" prior to starting the engine. I'm in annual fighting a power issue so it will be a bit before I can test the advice.

I agree with the call outs previously made, but I would also add my appreciation for the vertical trend. Call me lazy, but I can dial in my trim a lot faster and have it locked in a fraction of the time. 

It would be nice if the wifi was available for these updates when they get released.

Just came back from a flight and tried keeping my master switch on during my pre-flight to let it "warm up". In total, it was on for 13 minutes before I got in the plane and cycled the master switch. Unfortunately, it was very turbulent up here in the northwest and my compass was useless as a baseline. Instead, I did a few turns and compared my heading to gps track (not trying to match them, but having consistent disagreement from turn to turn). It seemed to work! I don't think I'll be able to give it another go before next weekend but today's flight was promising.

  • Like 1
Link to post
Share on other sites

That sounds very promising.

Thinking through the logic. I wonder if the fix is centered around the instrument being warmed up or if the engine vibration is affecting the initial alignment.

Link to post
Share on other sites
  • 1 month later...

     Do the other folks’ that have the AV30’s installed have the units tied to the avionics buss or main buss? Mine are on the avionics switch and the Uavionix engineer suggested they need to be on the main buss, that way they will be on with the master switch and remain on during engine start after alignment. I’ve tried the procedure suggested, power them up, wait for alignment, power avionics off and put the -30’s in battery mode for engine start. Did not seem to help. AI still lags in roll in flight showing a bank when level and level when banked. Definitely not safe for IMC. I am getting close to pulling them and trying the E5. Anyone out there have a recent install or PIREP on the latest version? I just want a reliable AI and DG. Funny but I had that with the original vacuum system.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.