[stu517]

looks like a good start. i would hope that you would give yourself some backup instruments (Attitude, Altimeter, Airspeed). I have flown behind a number of the newer EFIS's both certified and experimental - they all have issues and the manual backup is well worth the $$. good luck

 

[lostpilot28]

looks like a good start. i would hope that you would give yourself some backup instruments (Attitude, Altimeter, Airspeed). I have flown behind a number of the newer EFIS's both certified and experimental - they all have issues and the manual backup is well worth the $$. good luck

Hi Stu,
No mechanical backups! This is what it will be, except maybe the addition of an IFR GPS. Each EFIS can act independantly from the other, or in a Master/Slave config so you have synchronized data. If one unit goes down, the other takes over. It's a beautiful setup, really, and I'm not inclined to move backward in technology with my panel (unless I have to!).

 

[pierre smith]

We too....

Hi Stu,
No mechanical backups! .......... It's a beautiful setup, really, and I'm not inclined to move backward in technology with my panel (unless I have to!).

....will not have any round gauges on Mr. Blanchard's -7 we're helping build. Only dual Dynons with the new terrain and synthetic vision/autopilot combo.

Each is considered a redundancy for the other and have internal backup batteries in case you lose alternator/battery power.

Makes for a really neat install.

Regards,

 

[SteinAir]

Sonny - Nice panel!
Pierre - you guys should also end up with a nice panel.

Quick question for both of you....are you both considering these setups to be used for IFR/IMC flight/approaches, etc?

This truly isn't anything other than to settle my personal curiosity, but do both of you feel comfortable using MGL and/or Dynon screens (or any mfgr for that matter) as the only equipment in the panel as suitable for IFR or IMC flight with no "2nd system"? Basically what I'm asking is that would both of you go barreling off into the overcast or clouds with a singular system?

Just curious.

Cheers,
Stein

 

[pierre smith]

Yes...

Sonny - Nice panel!
............., but do both of you feel comfortable using MGL and/or Dynon screens (or any mfgr for that matter) as the only equipment in the panel as suitable for IFR or IMC flight with no "2nd system"? Basically what I'm asking is that would both of you go barreling off into the overcast or clouds with a singular system?

Just curious.

Cheers,
Stein

Stein, when I consider how often we used to fly a Cherokee or Cessna xxx into soup and do approaches with a vacuum system and the only redundant instrument was an electric turn and bank Yes, with dual Dynons/GRT/AFS/Oddysey,etc, I would be very comfortable after I fly with a buddy as safety pilot. I mean, two full panels, dual engine monitors, each capable of standalone ops, regardless of alternator failure.

Regards,

 

[Ironflight]

I think that what Stein was asking was really about "unlike redundancy". While I am a proponent of thinking outside the box on what we really need in our panels, and I am quite comfortable without "traditional" instruments as backups, I do want to protect from any single point failure. In the old days, the electric T&B protected us from a vacuum failure for instance> But if you have two of the same electronic things, are you fully protected? What happens if there is a software failure? A bug takes them both down, and you have nothing. This has happened more than once in certified and military systems. I solve this in my airplane by having a separate autopilot design from my EFIS. In Louise's -6, we have an ADI Pilot II as backup to the dual Dynon's. I personally wouldn't be comfortable with a single-source panel, because bug-free software does not yet exist. The truth is that the hardware you have installed in an EFIS airplane is actually quite unlikley to fail. Your biggest potential for trouble is the software, or an electrical fault.

Unlike redundancy is your friend!

Paul

 

[SteinAir]

Stein, when I consider how often we used to fly a Cherokee or Cessna xxx into soup and do approaches with a vacuum system and the only redundant instrument was an electric turn and bank Yes, with dual Dynons/GRT/AFS/Oddysey,etc, I would be very comfortable after I fly with a buddy as safety pilot. I mean, two full panels, dual engine monitors, each capable of standalone ops, regardless of alternator failure.

Regards,

That was my point exactly. If you only have two identical EFISes in your panel (and nothing else, not even a handheld GPS)...when one is leaning left and one is leaning right - which one is correct?

Paul eluded to this, but obviously the heavy iron requires a "non like" system as a backup. Different technology, different sensors, etc..

You could have 10 EFISes, but if they all rely on pitot/static or gps or other singular external aiding to keep the attitude display correct, it does you little good to have all identical systems if they can be brought down by one bug, pitot/static, gps, or whatever.

Anyway, don't want to start a war here - just making sure people are really thinking this through. I'm pretty sure the EFIS mfgrs would never openly market their systems as a sole source of primary and simultaneous backup for IFR/IMC flight....and if they did they obviously haven't talked to their lawyers!

Cheers,
Stein

 

[Radomir]

But Stein.. let's review the facts.. 90% of RVs seem to be equipped for IFR (some for "light IFR")... yet... of those... what.. maybe 2-3% (tops!) ever see a whif of cloud.. seems like most of that small percentage seems to be backed up "ok"

My biggest concern is to have redundant shade for the sun in case the cheapo walmart one pops off the canopy and doesn't wanna stay there

ok.. couldn't resist.... sorry

 

[N941WR]

I agree with Radomir about the cheep-o K-mart shade. I hate it when it falls on my head!

As for duel, indipendent but identical systems, hitting the same bug at the same time, in the same plane...

IMHO, that is very unlikely, possible, but unlikely. Even when dealing with the poorly written business systems I've managed, it was very rare that two "identical" systems whould hit the same error at the same time.

Granted, the same inputs should cause the software to react exactly the same every time. This is true, but when the inputs are created from an aircraft flying through space, it is unlikely two units sitting side-by-side, running off their own internal clocks, would hit the same bug(s) at the same time in a well tested application.

 

[Ironflight]

But Stein.. let's review the facts.. 90% of RVs seem to be equipped for IFR (some for "light IFR")... yet... of those... what.. maybe 2-3% (tops!) ever see a whif of cloud.. seems like most of that small percentage seems to be backed up "ok"

well, you're right Rad - the vast majority of flying I do in my well-equipped IFR bird is VFR - I rarely have to file for a number of reasons, but it is nice to know that if I DO have to file, I can do it with a significant amount of risk reduction already in place.

I am a lousy gambler - I have no tolerance whatsoever for losing, so I simply don't play anything but a sure bet. And that's just with money - when it's my own little pink body, I make sure that everything is way stacked in my favor.

You're right that the likelihood of having a failure on the one day that a person decides to file IFR in their airplane is small. However, the severity of consequences if it id happen are very high. I like to analyze risk on a two-dimensional scale, with severity of consequences on one axis, and likelihood of occurrence on the other. I can live with points close to the origin - low probability and low consequences. It is when you start getting out to the high end of "bad consequences" that I personally get nervous.

All just food for thought. I happily flew IFR with little to no redundancy decades ago because we didn't have any options. Today, we have very cost-effective options for dissimilar redundancy with digital systems. Many manufacturers stuff talks and plays well with other systems from other companies, so it is not hard to put together a system with fault tolerance and great capability.

Paul

 

[pierre smith]

What I was thinking too..

As for duel, indipendent but identical systems, hitting the same bug at the same time, in the same plane...

IMHO, that is very unlikely, possible, but unlikely. Even when dealing with the poorly written business systems I've managed, it was very rare that two "identical" systems whould hit the same error at the same time.

....that the law of averages should be on my side. I can see Paul's not using a Dynon autopilot driven from a Dynon EFIS tho'. It folds when the Dynon (or whatever) does.

Regards,

 

[flytoboat]

stuff happens

Even Big Buck military avionics can have problems:
http://www.freerepublic.com/focus/f-news/1791574/posts

 

[RV10Rob]

What do you use for a tiebreaker? If one box goes blank, it's fairly straightforward--use the backup. What if your EFIS shows a left bank and your ADI or other backup shows a right bank? In this situation it'd be nice to have an independent autopilot, but since I'm installing Dynons, I won't have that option without a significant increase in expense.

-Rob

 

[vic syracuse]

I think I covered some of this in my post regarding electrical system redundancy. I agree with Paul, and I practice it, that 2 EFIS's are OK without mechanical backup if they are from different vendors. I have dual Cheltons, a Dynon, and a Trutrak autopilot. I can fly the airplane independently with all 3 of them. The only single point of failure is the pitot tube, but it is heated, and I decided that 2 pitot tubes would probably be overkill on an RV.
Just this past week end, on a trip to Kansas City, the PFD screen of the Shelton rebooted in flight for no apparent reason. The second screen took over as advertised, and immediately flipped to PFD from MFD. The reboot takes a while (2-3 minutes). That's a very long time in IMC. I was VFR on an IFR plan, so no real problem.

Vic

 

[allbee]

I made my panel with both. Yah, it's nice to have backups, but that's not why I did it. When I fly I like to look over and see something on a round dial. My brain computes it much faster than a digital read out. When I do things through the day I look at a digital clock and stop and think, even a second is too long. But when I look at a round clock, it's registered right away, I don't think about it. Same in the airplane, when landing I know where that 70 is, I see it on the round guage and it is computed right away, on the digital, it will be a blank stare for a second, something I don't want. More than likely I'll be putting a little white dot next to the 70 for landing. Than I can have an even shorter look without thinking.

 

[vic syracuse]

Steve, I also found the switch to all digital quite a challenge. Especially when it comes to maintaining altitude. It was easy to see trends on the old round VSI. It ppointed up or down, and the hands on the alitimeter usually moved first. With digital, it does take a few brain cycles to compute. I like the "trend" indicators on some of the EFIS's, both for airspeed, altitude, as well as the flight path indicators. They become very helpful over time once you get used to them.

Vic

 

[Ironflight]

The F-22 crossing the dateline referenced above is the exact type of bug that I am worried about - it is induced by external "conditions" acting equally upon all internal devices - software design has gotten to be so "big" that rarely do you have one person that understands the entire system (including how it interacts with the outside world), and therefore, even though every designer is sure that their portion is bomb-proof, the overall system reliability might not be. I have experienced this many times in real space vehicles, and am therefore a believer. You go from being 100% confident in your equipment to 0% confident in an instant. So pick your systems so that it is unlikely that they will all go belly up at the same time.

And yup - a man with two watches never knows what time it is...unless he uses a little common sense, and a third input of "some" kind....

Paul

 

[lostpilot28]

Sonny - Nice panel!
Pierre - you guys should also end up with a nice panel.

Quick question for both of you....are you both considering these setups to be used for IFR/IMC flight/approaches, etc?

This truly isn't anything other than to settle my personal curiosity, but do both of you feel comfortable using MGL and/or Dynon screens (or any mfgr for that matter) as the only equipment in the panel as suitable for IFR or IMC flight with no "2nd system"? Basically what I'm asking is that would both of you go barreling off into the overcast or clouds with a singular system?

Just curious.

Cheers,
Stein

Thanks, Stein. For now I'm only a VFR pilot...and I've only given a little bit of thought to what I'd need for making my RV an IFR capable machine. On my original post I noted that I'd probably only add an SL-30, but after hearing Paul's comments on the value of a GPS Comm for IFR, I decided that it would be a better option.

Again, listening to your thoughts, and Paul's I think I would probably add some redundancy in somewhere. My original plan was to have a Trio Autopilot in the middle of the panel...that would've taken care of the EFIS's going down. But, this of course is my line of thinking with virtually no flight time behind the MGL units. From everything I've read and heard of, they're very solid, stable systems. If I'm completely comfortable with their performance after a couple years and I get my IFR ticket, maybe I'll feel I don't need a backup. Not that it wouldn't be a good idea, but like Pierre said it's the law of averages leaning in my favor. Of course, Paul countered that by saying the risks of being wrong are very high.

Lots of food for thought. I guess I don't really know what I'll do down the road.

 

[SteinAir]

Wow guys...this has and is an unusually respectable discussion here...I'm quite enjoying reading everyones take on this.

I guess my end standpoint is that I'm with Paul. I'm really not concerned about the hardware much at all, therefore the multiple electrical systems and battery backups while nice aren't what I mean by redundancy. I'm not a rocket scientist like Mr.Dye, but I have spent a lot of time around a lot of high end stuff in the heavy iron world. The truth of the matter is that you're far, far more likely to have a software bug/glitch/gremlin/failure than you are to have a hardware or electrical failure.

Without getting into the argument about who's software is better, etc.. let's just say that my personal opinion is that I wouldn't be in the clouds with a single source instrument as my only attitude reference....like I said, doesn't matter if you have 30 of the same ones if you end up losing a single parameter or two that those systems depend on for their integrity. As pointed out before, it has, can and will happen....so like Paul said (and my take is) design your system so everything can't fail at once. ANY setup be it engine, avionics, electrical, etc.. that has a single point of failure is not good ju-ju especially if you intend to fly IMC.

BTW Sonny - it appears that you have a good head on your shoulders and have reasoned your choice through. I can't argue with that, nor find any fault in it. Your craftmanship also appears to be quite good (at least from what I can see)!

Again, nice to see a civil discussion where we can all share our opinions without getting nasty! Keep it up

My 2 cents as usual.

Cheers,
Stein

 

[digidocs]

Paul and others have raised an interesting point about backup attitude sources while IMC. I'm curious what is supposed to happen in this scenario:

I'm in turbulent IMC in my SR22 equipped with its Avidyne PFDs and a backup vacuum attitude indicator. Suddenly, I notice that my PFDs show a left bank while my mechanical AI says I'm going right. Whiskey compass is bouncing all over the place. Vacuum looks good. What do I do?

In a six pack panel, the third gyro can break the tie. However it seems that some FAA certified ships only have two "gyros".

Hmmmn.
-DC

 

[Ironflight]

I'm in turbulent IMC in my SR22 equipped with its Avidyne PFDs and a backup vacuum attitude indicator. Suddenly, I notice that my PFDs show a left bank while my mechanical AI says I'm going right. Whiskey compass is bouncing all over the place. Vacuum looks good. What do I do?

In a six pack panel, the third gyro can break the tie. However it seems that some FAA certified ships only have two "gyros".

You're in at least as good a shape as you were in the old days with a vacuum ADI and DG plus an electric T&B - the FAA assumes you will have ONE failure, and that you will recognize it (The wisdom of those requirements are left as an exercise for the student...). The certification requirements haven't changed, and probably won't but you can do better in your Experimental, because you can....

In the Cirrus? I don't know the systems well enough to have a meaningful opinion I'm afraid....

 

[SteinAir]

Hi DC,

In your case (certified) this is a wholly different scenario. The processes that your Avidyne unit had to go through for software (and hardware) certification are EXTREMELY extensive in comparision to what our experimental counterparts do (same with Garmin, Chelton, Crossbow, etc..).

There is a lot of processing going on inside your box where it tells itself if it's sick or ill or otherwise not accurate. There is a whole plethora of things being monitored simultaneously to make these deteminations. I'm sure Paul or Rob Hickman can go into more detail on this, but belive me it's very complex. For example, if the certified systems miss even a few bytes / packets of info from the AHRS to the main system, it has to immediately tell you that and is supposed to put a big red "X" over the affected unit. This may only be temporary while the system figures out what the sickness was, but at least it knew it was ill. Here's a quick example. If the airplane is showing a left bank and in reality is not banking left (flying level or turning right), your certified system will know that, because it's watching a bunch of other parameters as well (GPS, Air Data, Magnetic Data) etc.. If all of those parameters don't agree 100% that the maneuver being depicted can't be matched through all the various paramenters simultaneously, you get a failure. The AHRS boxes in the certified stuff are also quite different (and read expensive) as compared to what we see in our affordably priced experimental stuff. Not saying they are worse or better, but I will say there are a whole lot more sensors in the certified gyros than you see most of the experimental ones.

Anyway, there is a HUGE list of differences between the certified EFISes and the experimental EFISes, especially when it comes to software, AHRS, etc.. (and most of all cost). Some of it so technical it wouln't make sense to most of us. Does this make our low cost experimental systems bad? No, it just means you really need to do your homework before you rely on them 100% for your life in IMC with no backups and try to understand you just can't compare them as apples to apples...

In the end, your scenario is certainly possible, but highly unprobable just due to the systems you referenced and how they work behind the scenes. Even though many EFISes now are starting to look similar on the screen, what goes on behind the screen is startlingly different....and remarkably different in price!

My 2 cents as usual.

Cheers,
Stein

 

[stu517]

avidyne backup

lucky for you guys I actually have a few hundred hours behind the avidyne and have experienced failure in imc. what you have is a 3 pack on the bolster (backup #1)and and an autopilot with an independent gyro (backup #2) (stec 55x) so it is pretty simple - if the avidyne fails (they do a lot more frequently than they should) you are instructed to pull the breaker on the PFD. This stops any interaction with the autopilot vis a vis heading info. you use the 3 pack on the bolster for main instrumentation - attitude, altitude, and speed. keep the autopilot engaged if possible - flip one garmin to get track and speed info. so the PFD doesn't affect your ability to let the plane continue flight on autopilot. with WAAS you ask the controller for a GPS/LPV approach instead of the ILS and essentially the plane flies you to 200 ft. BTW running these scenarios is part of annual recurrent training that most cirrus drivers take. i have a merciless cfi that fails just about everything so my last approach is handflying using a garmin 430 for primary nav.

I have several hundred hours of actual and 100+ approaches. I know from experience that it doesn't take long to get yourself wrong side up in a jam - and believe me you have no spacial reference. As "uncool" as it might look, I get comfort seeing a 3 pack backup just in case. just my 0.02

 

[SteinAir]

Good points....Avidyne certainly has it's share of issues, but my main point was that if it fails, you know it. The question about the Cirrus was posed in a way as to what to do if one leans right and one leans left...having two devices and not knowing which one was failed. My point was that whether or not the Avidyne is good (or not - I won't go there) you should at least know when it fails. Typically with certified systems they do not go about their merry way if/when they have a problem...somehow the pilot will figure it out. Good to hear someone has "been there done that" and can give us a first person account....and is hear to talk about it. Thanks for that input - your point about the autopilot with an independent gyro is well taken, and one I forgot to mention. I usually tell my IFR customers that the first backup is and should be an autopilot! Autopilots don't get nearly as disorientated as our brains/eyes/inner ears do.

Cheers,
Stein

 

[Jordan1976]

For example, if the certified systems miss even a few bytes / packets of info from the AHRS to the main system, it has to immediately tell you that and is supposed to put a big red "X" over the affected unit.

Yep, we do that too, we just annunciate it in a different way (black and white screen). CRC checks on communications is pretty basic stuff.

Not saying they are worse or better, but I will say there are a whole lot more sensors in the certified gyros than you see most of the experimental ones.

http://www.xbow.com/Industry_solutions/gyro_guide.aspx

Says that Crossbow uses three MEMS gyros, three accelerometers, and three magnetic sensors. Just like we do, and just like every other ADAHRS in this market. No mention of extra sensors for cross-checking.

Just like a man with two watches, more sensors inside a box doesn't always give you more information. The more sensors you require, the more likely one is to fail.

...your certified system will know that, because it's watching a bunch of other parameters as well (GPS, Air Data, Magnetic Data) etc..

Interestingly, Crossbow (and the vendors that re-sell Crossbow) claim that they do not use anything external. No GPS, no pitot, no magnetometer, and this is something they are justifiably proud of. So what exactly are they cross-checking against? Apparently cross-checking is not a requirement for certification.

Garmin's G1000 ADAHRS requires GPS or pitot/magnetometer to work properly. Not to cross-check. To even work. They warn you that if you lose GPS, they'll keep giving you attitude, but it's not as good as normal. That's not cross-checking, it's assistance. It's all right in the user guide. They use this data to actually assist them in coming up with a solution at all.

Certified AHRS units aren't really that different than the non-certified ones. If we thought adding 2 of every sensor would make people safer, we'd do it. Plenty of certified AHRS units have bugs that have caused them to be non-flight worthy for long periods of time while they work on them. Garmin G1000's have been known to lock up. Aspen EFIS units have been known to tumble without any error reported. As the gentleman above mentions, Avidyne units go wonky a lot, and are you 100% sure that they annunciate it every time? It's hard to tell the user the AHRS is iffy when all of Windows NT has locked up beneath you. Has even a single customer here ever had a Dynon EFIS lock up hard on them in flight?

Apparently all that software "certification" doesn't actually do that much. Last I checked, certification has a lot more to do with the way software is documented than how it actually functions. I've never seen a TSO requirement that AHRS units have levels of cross-checking. People should read the actual TSO certification requirements before assuming certification means "fully tested and can never go wrong or lie to you."

My two cents too.

--Ian Jordan, Dynon Engineer

 

[Will_S]

I love the look of the glass cockpits and can't wait to fly them in my RV. But personally I wouldn't fly just glass in my RV, I will have analogue AH/DG/Alt/Nav as backup regardless of how many screens I have. Just me I suppose, but flying is about reducing risk and that is one risk I feel I need to reduce. IFR instruments become really important when trying to find the right bit of ground at the end of a flight - its interesting to see how little instrumentation you can safely finish a flight with if you retain proficiency and practice.

Stein, regarding 2 instruments disagreeing with each other: I have had experience when flying a twin cessna single-pilot in solid cloud my AH disagreed with the co-pilot side AH, not total failure - but one of them was on the way to total failure. I had to choose one to believe, knowing that if I chose the wrong one I would see it pretty much instantly on other instruments (ASI/Alt/T&B etc). I did choose the wrong one at first and found airspeed increasing, altitude descending, heading changing - didn't take long to work out which was right. Passengers never noticed. By the time I landed after an approach (not to minimums) my AH was dead.

Its not really a problem without a solution, at least you have a 50% chance of being right - so long as you are aware of signs you picked the wrong one and you transition from one to the other gently - no sudden moves. A dead analogue can be harder to pick than a dead screen, it may just stop moving.

I know the maxim for IFR flight is "believe you instruments" but I prefer sceptical belief - if they prove themselve right I will believe them for the moment.

For that, and other reasons, I will ensure I have backup basic analogue flight instruments and nav and hope I never need them in earnest.

 

[Rainier Lamers]

Anything can fail

A good friend of mine had one of the most reliable airspeed indicators on his open cockpit trike. A transparent tube with a little ball lifted by air pressure. One day he laughs and tells me of his in-flight ASI failure. A bee had aimed straight for the little feeder tube and smashed itself in there solid - ASI kaput !

Point is, anything can fail. Does not matter if it is our EFIS or a top certified and qualified cream of the crop simple on/off switch. If it can break, it will. It's just a matter of time.
Not one of the current EFIS systems will last forever. None. Zero. How long ?
5 years ? 10 years ? Maybe longer, but it will break.
But the EFIS has become (or is becoming) consumer goods. It's getting low cost and soon it's going to get even cheaper. It becomes replacable. You might even want to upgrade to the latest just for the fun (we are builders, right ?).

Now having said that, as EFIS designer and builder, I tell anybody that cares to listen to fit a backup. What backup ? Depends:
What do you need to fly your aircraft safely at a minimum ? What do you need to complete your mission if your panels are dark ?
Answer that and you have your answer. Surpise - everybody will have a different answer. Why ? because everybody does different things with his aircraft.

In a simple scenario, a backup ASI and handheld GPS is likely all you need in VFR. If you need to do this in solid IMC and have the required experience and training, you need a little more.
That is it, no more, no less.

Oh yes, don't forget a few other important items: Backup engine (you never know...), set of spare wings, tail feathers, gear,...
Point is, things break, anything can break.

Your EFIS is perhaps low on the list of priorities when you consider the above.

Rainier
CEO MGL Avionics

 

[Rainier Lamers]

The AHRS boxes in the certified stuff are also quite different (and read expensive) as compared to what we see in our affordably priced experimental stuff. Not saying they are worse or better, but I will say there are a whole lot more sensors in the certified gyros than you see most of the experimental ones.

Stein,
I have to throw a trip line there.
That is not true in all cases.

As example I have to list our SP-5 AHRS. It is not certified. It is also a huge amount cheaper than the certified Crossbow AHRS-500 which is often used as a benchmark.
It uses the SAME IDENTICAL gyros (designed by former British Aerospace) and the very latest in accelerometers that simply did not exist when the XBow was created.
Like the AHRS-500 the SP-5 performs unaided and its algorithms, even though different are at least as good.

Certification is an arduous and expensive task and due to our remoteness to the U.S. something we are unlikely to do. TSO-C4c is now quite easy to pass with MEMS sensors - the biggest issue is DO-178 which effectively is about documenting the software design and design process (which is not very compatible with the way things are done in real life in small companies).

The point is, certified does not mean better. It "can" mean better but that it as far as I am prepared to take it.
A good, small and nimble engineering outfit that does uncertified CAN have the edge. It can move much faster, use unrestricted methods, tools and parts as deemed best fit. It does mean it has to be satisfied with a tiny (and crowded) market - nothing comes free. But it certainly can produce better goods if it can and wants to. Of course the reverse is also true.

Now here comes an interesting statement:

In a way an outfit that produces uncertified goods still gets certified. It just does not get certified by the FAA. It gets certified by the actual users of the system and their approval or disapproval. It is also subject to a quite powerful peer review system, unofficial as it is. All of this made possible by the internet and forums like VANs.
Everything made is subject to intense scrutiny by a small army of very knowledgeable people that have their hearts set right in the aircraft building (creation ?) process. Not a couple of semi-skilled bureaucrats. Real experts. Real and meaningful judgement. This is what counts.
Trust me on this one: no designer and manufacturer of an uncertified piece of gear can afford to make something that is bad or dysfunctional. If he does, he's out as quickly as you can type http://
Now, even that is better than what the FAA can do...

Rainier
CEO MGL Avionics

 

[pierre smith]

My six pack looks better all the time...

.....since I just read in Sport Pilot last night about how needles/hands on round dials are easier to read.

In race cars, the engine gauges are often turned so that all needles are straight up when things are normal. A quick glance is all it takes and if a needle is displaced from vertical it requires more scrutiny. That's when the gauge is actually "read".

When I first started flying the -4 with a Dynon 180 it was the altitude that was more difficult to hold because you're forced to read the digital display instead of just glancing at the hands of an altimeter without actually reading it. I found that the little lavender color bar that indicates vertical speed easier to use....

I was gonna "upgrade" my -6A to glass, now I'm not so sure anymore....

This has been very eye opening and I thank all of you as well.

 

[the_other_dougreeves]

.....since I just read in Sport Pilot last night about how needles/hands on round dials are easier to read.

In race cars, the engine gauges are often turned so that all needles are straight up when things are normal. A quick glance is all it takes and if a needle is displaced from vertical it requires more scrutiny. That's when the gauge is actually "read".

Human interface design is a very puzzling field to me. We learn something and then quickly forget it because the designer / engineer comes up with somethine new or the marketing people want to change the design so they can sell it as a newer, better design.

Example: I still use a IBM Model "M" keyboard. The Model M was made by IBM in the days when they still made PCs. It has outstanding tactile and auditory feedback - you have no question about when the key is pressed and it requires a certain amount of pressure to actually press the key - very much unlike the modern keyboards of today that all feel like pieces of gel. You can look away from the screen and easily type, knowing that you've pressed each key.

So, steam gauges vs electronic displays. Yes, dials are easier to glance at and read quickly. All the better if they are alligned so "up" is normal.

However, the power of electronic displays is in information density and the ability to change the display when something is abnormal. On the first, when you're flying IFR and trying to keep a scan going, having the information in a smaller area makes the scan easier. Modern PFDs combine the ASI, ALT, AH, DG, CDI, GPWS/terrain function and engine instruments all into a 6" screen. Does this help you when you're flying VFR? Not as much. What you need then is the ability to quickly glance at the instruments and read them, and also to recognize trends.

On to recognition of abnormal values ("unusual modes" in process control). If a value, let's say oil temp, is either increasing too quickly or is too high, a gauge just sits there and points. An electronic display is able to (1) change color, (2) flash, (3) grow larger or reposition the abnormal reading and (4) sound an auditory alarm. This helps the operator (pilot) identify the unusual value and figure out what to do. In my mind, electronic displays are clearly superior to dials for airplane engine monitoring applications for exactly this reason.

We chose steam gauges for our CT because we would be flying VFR only, and I'm happy with that decision. I wish I had the Dynon EMS, though.

TODR

 

[RV7ator]

Everyone's talking failure. What about the Black Screen of Ambiguity?

I've experienced multiple instances of D-10As putting up a black screen, usually - but not always - from overtemp. Is it still valid info, or not? Dynon says, "Oh, it's o.k. It's signaling that some element of the guts is exceeding mfgr's spec." That is no help at all. Who you gonna call?

Why, my round gauges, of course! As others have noted, they are immensely easier to interpret than reading a digital display, plus, being stone simple (the instruments, not me!), I trust them if the EFIS disagreed.

Dynon users: vent your panel. The last black instance was during holding while awaiting vectors for an ILS. The sun, warm ambients, and low speed overheated the poor baby. Only because of the back-ups did I continue.

Two-Dynon users: consider that their Smart Buss requires configuration, something you might be loath to redo in troubled flight to get around a failure that takes down the buss.

Multiple, several redundancy is best. How much you want for whatever percentage of your time in the clag depends on your (let's be real) IFR expectations.

John Siebold

 

[Jordan1976]

I've experienced multiple instances of D-10As putting up a black screen, usually - but not always - from overtemp. Is it still valid info, or not? Dynon says, "Oh, it's o.k. It's signaling that some element of the guts is exceeding mfgr's spec." That is no help at all. Who you gonna call?

This is exactly the situation Stein was talking about. We tell you when there's a failure. That's not the exclusive arena of certified units. Our theory with the black and white screen is that if all you have is the EFIS, it's better than nothing. But if you have backups, use that.

Two-Dynon users: consider that their Smart Buss requires configuration, something you might be loath to redo in troubled flight to get around a failure that takes down the buss.

All Dynon EFIS units in a panel show local EFIS data. So if you have two Dynon EFIS units, and the buss blows up or one unit dies, the second unit just keeps on going. Sure, it throws an error, which you can acknowledge with one button. Not sure why you'd need to reconfigure in flight.

 

[Brantel]

Not sure why you'd need to reconfigure in flight.

Ian,

Can you fill us in on Dynon's thoughts about if your EFIS that dies is the DSAB master and controlling the Dynon AP and or responsible for gathering the NAV/COM/GPS data for the panel, what then? Sure with a dual EFIS system you have redundant EFIS but what about a working AP or display of your EHSI?

Right now with today's products, I see a IFR Dynon solution as follows:

1 D180, configured as the master most of the time displaying efis and HSI
1 D100 most of the time displaying engine data and info pages
Backup AS, TT ADI, and ALT
Dynon AP, dual axis AP74 or 76
HS34
NAV Source and or GPS source of your choice

If the D180 dies, you loose your primary EFIS, AP, Nav/GPS data, and engine data -- unless you can reconfigure the DSAB bus on the fly????
If the D100 dies, you loose your redundant EFIS
If both die, you loose Efis data, Nav/GPS data, engine data, AP and must resort to the steam guage backups and looking at the built in displays of the NAV/GPS gear all while hand flying the aircraft in IMC.

 

[N55CU]

All Glass

I love my EFIS (AFS 3400) but I use my round gauge airspeed and rate of climb a lot as it only takes a glance versus a focus to get it from the EFIS
Randy Utsey
N55CU
RV-7
Charlotte, N.C.

 

[stu517]

Ventilation

A very important point was made here. Most of the current GA panels do not have enough cooling to support the increased amount of electronics being used. These all in one EFIS's generally put out a ton of heat and are sensitive as well. I fried two experimental EFIS's myself. Once I was able to adequately move air behind the panel things worked a ton better. BTW, this applies to certified planes as well. One of the "tricks" with Cirrus owners is to add a fan behind the panel when an annual is done. Best $$ spent. A good way to tell is to touh the metal casings of your avionics while on a long cross country. If they are hot to the touch you need more cooling. Any of the parts houses have them well under $200.

 

[Dynon]

Brian,

We've never sold the AP as having redundant sources. Buy a AP from any other company, and if it dies, it dies, just like we do. So think of it more this way:

D180 = ADAHRS, HSI, AP
D100 = ADAHRS

We think this is a very usable solution. As you say, you still have the displays for the navigational stuff right on the GPS display, so in the failure of the D180, you can still stay level and navigate.

Given that many people in the past configured this way:
D180, or D100+D120
Other AP

You still have the same amount of redundancy you always had, except you actually have a backup ADAHRS, so you're even better off, and you've saved some money. We think that two of our EFIS units, with one acting as the autopilot in your system (in other words, you consider one primarily as your autopilot, and the other as your primary set of flight instruments), will prove to be more reliable than the AP function as any standalone AP.

If you are really concerned about a backup HSI, you can wire a switch to the backup D100 that will shunt the serial data to it in the case of a failure. Then your backup D100 is still a EFIS and HSI. It is never an AP without ground reconfiguration.

 

[Brantel]

Thanks Dynon,

I just want to be 100% sure that I understand the limitations of the Dynon solution since my panel is 99.9% going to be based on the current Dynon products and the new AP .

That being said, I know all systems have some sort of limitations, it is just important to know what they are.

As long as I know the limitations, I can plan for the surprises that get thrown my way...

 

[SteinAir]

Brian,

....Two EFIS units in the system are just as reliable on the AP function as any standalone AP.......

Come on now, let's try to keep the rhetoric down here and stick with the facts. What exactly are you basing the above statement of fact on? How many fleet hours of reliability data do you have on your AP in customer installations? At the moment, I was of the understanding that with the exception of beta testers, there is a sum-total of zero production customer installed units in the air?! Am I wrong here?

Just trying to keep the facts straight here. There is a big difference between historically accurate reliability data and marketing/sales rhetoric.

Cheers,
Stein

 

[Dynon]

You're right Stein. We've amended the post to reflect our full opinion more clearly .

I think you're splitting hairs though. The point is that from a system-level perspective, thinking of one of two of your Dynon EFISs exclusively as your Autopilot gives you the same sort of robustness as a standalone autopilot gives you. Ie - if your "autopilot" EFIS fails, you lose your autopilot. If your standalone autopilot fails, you lose your autopilot.

Our Autopilot is fundamentally our EFIS, so there is no reason to expect that adding the Autopilot feature will result in a product that is less reliable than the identical hardware that we've shipped thousands of times.

The servos are admittedly unproven in the market, but we've built them to be as reliable as other Dynon products, if that's worth anything. We do have thousands of hours between lab and flight testing on them, though.

 

[wv4i]

Use all the info avail?

>>>If you only have two identical EFISes in your panel (and nothing else, not even a handheld GPS)...when one is leaning left and one is leaning right - which one is correct?<<<

The above question has been raised several times in this thread. The answer is what's the heading doing, assuming the ball is in the center? .

To expand further, if one pitch is up, the other down, what's the altitude doing? VSI? Airspeed?

Basics.

Now if these readings are all coming from the same source, then that really could be a problem. As devices become interconnected, at what point are they really no longer separate sourced? Are you sure that they are really separate sourced? And, as a practical matter, separate sources will make indentifying and repairing of the offending unit/sensor much simpler.

For purposes of this discussion, the 737-800 has dual EFIS, dual Nav displays, redundant sources whose failure will cause an auto transfer in some cases, and a multifunction electric LCD backup attitude indicator with a/s, alt, loc/gs, integral. The latter will run from the battery bus, and uses pitot/static info independent from the primary EFISes. All in all, a fairly typical setup in the newer Boeings, i.e. 737/757/767/777.

But we're not operating airliners or spacecraft here. And most will not be routinely taking SINGLE engine airplanes into hard IFR, i.e. ice, trw, wx to mins, etc., I hope? Been there, done that, and don't recommend it. And it IS possible for anything to fail, no matter how well thought out and sophisticated. Where does reasonable prudence and planning suffice? Years ago, we had an MD-11 that apparently had two of it's three 3D orientation systems fail, and those systems are/were tied into a host of systems that generated envelope protection with automated responses of various modes, i.e. autothrottle auto engage, horns/bells/lights, etc. Suffice to say that without the true professionals at the controls, this would have been a widely publicized event. But this WAS an airline operation.

More recently, we had a 757 get down to battery power only. Apparently multiple "independent" failures appear to be the cause, but the investigation is ongoing. To expand, this means standby attitude gyro only, as well as a loss of numerous other systems that really make your job tough. The pilots literally both had their hands full in the flare. This WAS an airline operation also.

If electronics, integration and design of components is your interest, and/or time/money's no object, that's great. But, we need to not lose sight of the ball here, i.e. what most of us really need for a safe operation. As alluded to in the quote, the GPS generated flight "instruments" page in the handheld Garmins should be enough to keep you out of a jam, in the failure modes considered by this thread, as well as settle the different indications issue that may be raised by competing EFISes. If not, should YOU really be flying IFR, VFR at night, etc.?

Back to just being a reader, and thanks for all the useful info this site provides.

 

[BlackRV7]

For what it's worth

On my original panel I was using a Dynon as my primary flight instrument and had backup round gauge for airspeed, altitude and VSI. This was my first experience flying behind a flat panel display. After my brain acclimated to "seeing" what I needed on the Dynon, the only round instrument I continued to look at was my VSI. Upon redoing my panel, after the incident, I opted to go with a GRT HX as my primary, Dynon upgraded to D10A and Dynon EMS D120 and a separate Trio Auto Pilot. I have had a couple people email me mentioning I probably should have gone with one manufacturer for esthetics, I went with different for independent safety reasons. I sold all the round gauges. Here was my thought process. I went with the Aero Electric ebuss electrical system design. The Trio is the tie breaker. In case the GRT with TU, I still would have the Dynon, with battery backup, as an operating attitude indicator. If the Dynon went TU, I still would have the GRT as primary, if I lost engine info from the Dynon engine monitor, tied to the D10A, no biggie, all I'm concerned about is keeping that big fan out front turning. While I realize I could have total failure, a lot is going to have to happen at one time. You know, I kinda figure if all that happens, it's just my day to say: "it's been fun". I might go down, but I'll go down fighting as I still have that stick in my hand.

With all that said, one intregal part of my panel is my independent Trio. Also, I really believe the one round instrument I would reconsider putting back into my new panel is the VSI.

I agree with Stein here, this is one of the most civil flat panel/backup gauge discussion I have witnessed.