Second that. And, come to think of it, and seeing the extremely limited panel area, perhaps the breakers and a few more things could be located elsewhere, overhead maybe as they are in airliners, or on a sidewall? I remember Piper Cubs with relatively empty instrument panels but with lots of auxiliaries on the side walls – especially starboard, as the port side folds open.
I haven’t flown the onex, but do you need backups anyway? If in doubt, wouldn’t a simple phipsitheta do the trick? Better for aeros, too.
ortac wrote:
Your layouts with 2x MGL Xtreme EFIS look good. But preferably with them side-by-side or stacked. Can you fit one above the other in the centre, it looks tight?
Not possible to stack them on top of each other 
The space is extremely limited. Side by side will go, but then there will be no space for the Nexus other than setting it horizontaly, and I definitely don’t want that. I have this XPanel software that is really fun to play around with, and try all sorts of things.
ortac wrote:
The G5 only really comes into its own when integrated with Garmin air data, nav source, autopilot, etc so not sure it fits well with your MGL architecture?
Of course it fits best with a G3X touch (typically according to Garmin), but it’s an incredibly capable device on it’s own. Full stand alone capability and with it’s own 4h battery. It really is a true redundant primary flight instrument (including GPS and AI) backup for any system, and at 1200 US$ it’s almost too good to be true. (Maybe I’m just looking for an excuse to buy one )
ortac wrote:
In terms of backup, I think that a conventional altimeter+ASI is preferable over, say, a G5
I disagree. All sorts of things can happen to old fashioned mechanical systems. The G5 is the ultimate backup in my opinion, with it’s own power source. Even for a full analog system, it will back up every primary instrument on the panel. The problem with an EFIS is not that it needs electrical power, but rather the “all eggs in one basket” principle. If something goes wrong with it, it doesn’t matter what, you loose everything.
Peter wrote:
In the above layouts it seems OK however because all the switches, and therefore all the CBs, are more or less in one place. But I would still group the CBs together because there will be circuits which will need CB protection but won’t have individual switches.
They are not CBs. I use electronic CBs (a separate box tugged away), so the switch is simply connected with a red/green diode for each system one circuit is supposed to protect. Green means ON/OK while red means tripped. No light means off (the circuit). I haven’t really thought that part fully through, but I will have a switch/circuit for the USB, lights, the RDAC, radio/transponder, EFIS/EMS. One box only have 8 circuits. Several boxes can be stacked for a whole bunch more circuits, but I hoped I could get by with just one. There will be some additional switches for the different lights. My (LED) lights also has some fancy electronic built into them, and sort of duplicates the ECB functionality, but haven’t really figured out how it works yet. I also need a main switch in there somewhere. With the iEFIS, all configuration of these circuits are done there, and also the switches themselves can be “virtualized” in the EFIS (can use the EFIS instead of mechanical switches to turn on/off stuff for instance, and a tripped “fuse” becomes a warning on the EFIS). This also gives backup and redundancy a whole new dimension however, and I wouldn’t to it that way without two main buses/batteries and two independent EFIS’es (I think).
mh wrote:
but do you need backups anyway?
Jan_Olieslagers wrote:
perhaps the breakers and a few more things could be located elsewhere
LeSving wrote:
The places marked “USB” (in big red letters) are for USB
Duh. I should have studied the mock-ups more closely!
I disagree. All sorts of things can happen to old fashioned mechanical systems
OK but the chance of something “weird” happening to your mechanical instrument at the exact same time as an electrical/EFIS fault is tiny. Something weird happening to two EFIS at the same time is (slightly) more likely.
A mechanical instrument is undoubtedly “more dissimilar” than a second EFIS. Immune to some common mode faults, eg. overheating, over voltage damage, also immune to bright sunlight, etc.
But whichever way you go, the reliability and redundancy should be more than enough for this kind of aircraft. So I would just go with whatever you prefer the aesthetics of in the panel.
One thing I don’t like about xpanel is that I couldn’t find a way to display the dimensions of the actual panel to confirm that they are in line with the real thing.
Otherwise you seem to have covered all of your options.
I used AFS from the USA and wouldn’t do so again.
ortac wrote:
OK but the chance of something “weird” happening to your mechanical instrument at the exact same time as an electrical/EFIS fault is tiny. Something weird happening to two EFIS at the same time is (slightly) more likely.
Who knows what the reliability is? I have not seen any data that backs up anything in particular. MTBF numbers are non existent for light airplanes, and this includes certified aircraft. The only thing we can do is to minimize the chances of one single failure leading to complete disaster, and only some redundancy or backup can do that.
EFIS vs steam is not a meaningful discussion in terms of reliability, we have no data. But one single failure in a steam setup, is simply a nuisance due to independant instruments, while it is a “disaster” in an EFIS with no backup.
May be worth noting that some uncertified avionics were disallowed from the UK CAA IFR project:
This may also be of interest:
The above is about a year old.
May not matter to a Norwegian registered homebuilt whose permit includes full IFR.
Something weird happening to two EFIS at the same time is (slightly) more likely.
Massively more likely. Take for example a malformed PDF file which will crash everything that runs a particular PDF reader. Evidently this risk has been judged as low by the AOC ops who use two Ipads in the cockpit, but they use them mostly for document display and there is usually a Plan B.
What exactly is “an accepted screen layout” ? When thinking about the G1000 and all it’s buttons and layers of menus, I wouldn’t call that accepted by any standard. The G3X touch is very different. Likewise, the default MGL screens are confusing to put it mildly. With some training and familiarization, the G1000 is OK, and so are the default MGL screens.
I can get the concept of “accepted screen layout”, but are there any standards or guidelines for it somewhere? This would be useful, but it has to be more specific than a subjective “look and feel”. I know these things have been heavily investigated and researched by the military, but I have never seen any guidance or similar. Also, in the Air Force, “cockpit familiarization” is a huge thing for all pilots/aircraft and it’s not taken lightly.
Are you building an IFR aircraft, to be good for instrument flight?
