6 hours ago, Sims Smith said:

What is causing this? 

the yoke is constantly sending a gear down command rather than just once when it changes.

plugins/xtlua/scripts/B747.10.xt.gear/B747.10.xt.gear.lua becomes

local runningGear=0
function B747CMD_gear_up_full_CMDhandler(phase, duration)
    if B747DR_gear_handle~=2 then
        runningGear=1 --up
    end
end  
function B747CMD_gear_down_full_CMDhandler(phase, duration)
    if B747DR_gear_handle~=0 then
        runningGear=-1 --down
    end
end  
function B747CMD_gear_off_CMDhandler(phase, duration)
    if B747DR_gear_handle~=1 then
        runningGear=2 --off
    end
end
function B747_animate_value(current_value, target, min, max, speed)

in the next update

last couple of people to report this fixed it themselves, its better to only send it once, but not critical. 

 

I get asked this quite a lot, again today. Cold and dark startup should something look like this. the video on the card as the engine starts links to a RW video of similar, if you spot any important differences between the two please let me know.

 

Looks ASTOUNDING!!!

Did a test flight with yesterday's master branch commit with the CF6-80C from the FMC options. General things for the to-do list.

• With the GE engine, EICAS still shows EPR, which it should not do on GE engines (just N1 and N2).
• There are typos in some switch tool tips.

I started the engine with some trial & error, figuring out the correct bleed configuration along the way. Unless I'm mistaken, APU bleed air should get me enough PSI to start one engine, so this means that APU bleed + open isolation valve on the flow path to the engine that is to be started + bleed valve to the engine + packs off = enough bleed air for the starter.
However, this only worked on the tenth(?) try as the starter would not run, despite enough PSI in the system (as confirmed by the ECS EICAS page). After that, bleed from the running engine + APU bleed should be more than enough to start another engine, be it one on the other wing (with the corresponding isolation valve open) or the one right next to the running one. Again, this worked very spottily, with starters only running after multiple tries.
Not sure what I did wrong or if this is a bug in the starter or ECS logic or a timing issue with xtlua.

While starting engine 4, I got a hung start by setting the fuel flow switch to "run" first and only then pulling the starter. The engine would not spool past ~40% N2. After having switched off the starter and fuel valve again, the engine did not spool down for a very long time (and I can not if and how manual intervention was required). I assume that a starter generally controls the start air valve, hence my expectation for an immediate spooldown. If the core starter logic works as expected, I'd look at the engine friction value in PlaneMaker to make the engine easier to spool down.

With all engines started, I hacked performance data into the FMC (no flight plan though) and took off from KPMD. Initial altitude was set at 10000 and 240 KIAS. With the autopilot and autothrottle on, I made the 747 fly in FLCH and HDG mode. Tried HDG Hold and HDG Sel and everything worked. Crossing 10000, I climbed to FL 340 and tried some VOR navigation. Hacking PDZ (112.2) and a radial into the FMC, I got a signal alright, but, as I found out just now typing this, the 747 doesn't even have a NAV/VOR autopilot mode (see here). Very odd, but fair enough.

After that, I hacked a flight plan into the FMC (KPMD - PDZ - SLI - KSAN). Entering everything went well and I got a route on the ND. At some point, I decided I wanted a STAR. Or an approach. Or both. Or a STAR again. Ah, just an approach, STARs in the US are just too long. Anyway, switching approaches seemingly worked just fine. But flying the route down to the approach exhibited two issues:

• Flight idle on the CF6 is way too high, at 60 to 70% N1. I thought this was an autothrottle bug, but even with manual throttle control, I could not get any less. This makes it impossible to slow down. While flight idle is indeed a bit higher than ground idle on engines, it should not be >180% of ground idle. The 60 to 70% Nx figure would be more suitable for N2 instead of N1.
  If this is due to the "three in one" engine logic going from a single curve in the ACF file, maybe it should be time to consider to split all three engine types into separate ACF files wioth curves more suitable to the engine in question (factory thrust derating can still be done in the FMC).
• VNAV target speeds should automatically switch to 250 KIAS when approaching 10000 ft with a negative rate of descent. It did not do that in my case and I was still at 290 KIAS target speed below 10000.

Despite being in LNAV mode, the AP refused to fly the approach. Granted, I did switch approaches fairly often before that and probably confused the AP, but the approach waypoints were merrily ignored, which prompted me to fly the approach manually with a lap around the airfield below the 1000 ft cloud deck (the 747 is a great hand flyer, quite fun).

• The "positive rate" callout always plays with a positive climb rate and gear extended. While this isn't a problem on a normal approach, it may be inappropriate if you have to climb with gear extended for some reason. So I'd add a "ground contact" check and a locking logic to the callout. Basically: If ground contact then unlock callout, if positive rate and gear extended, then play callout, if gear up lock callout.

Speed management with flaps and gear extended isn't much of a problem despite the idle thrust issue, and after a very tight base and final turn that would have driven tears of joy to the most hardened navy pilot and aircraft disaster movie director, reverse and autobrake did their thing. Everybody survived.

My verdict so far is "it's getting there".

 

P.S: Since the cabin interior (seats + lavs + galleys) is a separate object, you could tie their visibility to the "PAX/FREIGHTER" setting in the FMC with a custom dataref and some Lua logic. With some custom cabin texture, this should do for a makeshift freighter interior.

Edited April 30, 20233 yr by Bjoern

37 minutes ago, Bjoern said:

P.S: Since the cabin interior (seats + lavs + galleys) is a separate object, you could tie their visibility to the "PAX/FREIGHTER" setting in the FMC with a custom dataref and some Lua logic. With some custom cabin texture, this should do for a makeshift freighter interior.

that should be done already, just set freighter instead of passenger in the simconfig-plane config

38 minutes ago, Bjoern said:

With the GE engine, EICAS still shows EPR, which it should not do on GE engines (just N1 and N2).

hmm, there is three avionics sets, Im pretty sure if the GE type has stuck all you get is N1 and EGT on upper eicas and N2 and something else (N3 maybe) on the lower,, PW and RR are both EPR.

just did a quick flick through liveries and changing this seemed to be working?

First lost track of this thought it was a WIP but bit of dligent searching revealed the release of the latest April 2023 Version with all updates for XP12. Excellent work!. No comments on how it flies, they had a reputation of being an 'old man's aeroplane' due to the high level of automation of systems etc and their basically quite benign or friendly habits - in other words a great aeroplane. I had been waiting to make this my go to Boeing 747 in XP12, seems I have been rewarded!

No problems in XP12 getting it installed, started and a quick take off. Lots of fascinating touches to now explore as I dig into planning performance etc., and actually doing some flights to get used to it. (Oh and dare I say it, no issues for me with AA or shimmering or anything like that either!). 

Edited June 4, 20233 yr by coastaldriver

When is the next release?

The current state of the project can always be downloaded from https://github.com/mSparks43/747-400/, using the "Code" --> "Download ZIP" button.

All in all impressive. Now have a hang of the systems etc and built in delays so getting it all up and running and flying not an issue.

The FMS is another problem for me altogether. I spent hours looking at the various videos of the system on GITHUB that purport to illustrate that this system should function just like the real thing did. If I try to create a flight plan or route on the fly I cannot get past a departure point and a destination and trying additional waypoints gets me the INVALID KEY response. IF I load a flight plan via fms it displays fine but how do I get a DIRECT TO function operative to get to  a waypoint along the plan route first. I found the AP flew the plan back to the departure point first then on to the planned route. Not sure how the automatic navaid tuning works (I can see some but not all) but how do I get the beast to fly and ILS or VOR or how do I get an ILS into the Plan or Route. I thought I had done that but it flew something I could not recognise or understand. Thoughts and advice appreciated. 

On 6/6/2023 at 12:15 AM, coastaldriver said:

Thoughts and advice appreciated. 

probably the biggest remaining compromise. RTE and LEGS pages still rely on the XP default FMC.

second biggest compromise is there is still no way for the rest of the systems to get speeds and full altitude data out of the default FMC flight plan.

On 6/6/2023 at 12:15 AM, coastaldriver said:

DIRECT TO

in XP11 when airborn the legs page becomes the XP11 INTC_DIR page instead of legs.

in XP12 you can just drop the waypoint you want to go to onto L2 on LEGS and execute it.

On 6/6/2023 at 12:15 AM, coastaldriver said:

beast to fly and ILS or VOR or how do I get an ILS into the Plan or Route

in APP select an ILS approach, NAV RAD page will switch from PARK to the autotuned ILS, you can also manual tune from NAV RAD by entering the frequency and course. those pages should be to spec.

Slight deviation from spec there is you need to be within 50miles of the ILS instead of 120, because searching for the ILS is a pretty heavy operation.

Edited June 8, 20233 yr by mSparks

Thanks i suspected this was the root cause or underlying problem - same issue appears all over for others as well due to the lack of updates-changes to the FMS and GPS in XPlane. 

9 hours ago, coastaldriver said:

same issue appears all over for others as well due to the lack of updates-changes to the FMS

yeah, been on the list since

https://github.com/mSparks43/747-400/issues/724

latest "step closer" my side to fixing it up went in to xtlua back in early march

https://github.com/mSparks43/XTLua/commit/ff7f947f7a5fd8d47df557e03d48278737bc274e

In many ways I prefer them as "Laminar problems", the 744 is getting a decent amount of flight time across the community, so being able to use it to identify and file bugs against the core sim means they can get fixed for all the others. Once it goes full custom any such issues become 744 problems. 

There is two or three other aircraft (maybe more now, I barely keep track) being developed off the 744 codebase, so eventually the flightplan related FMC pages are going to get the same kind of loving as everything else, but to do a good job of it I need more flightplan management stuff in the Xplane SDK than we have now, which isnt currently on anyones top priority list (and will also be XP12 specific).

Thanks for that information, confirms my suspicions re Laminar and the FMS-GPS issue. Given that it so fundamental now to onboard flight navigation and aircraft management I find the lack of attention to this major Laminar cluster peculiar to say the least. You get a great model nice EFIS or Electronic ala G1000 flight display and then limited functionality to the point that I actually go out of my way to find or fly in the sim at the moment aircraft that have the basic radio navigation kit (ILS VOR NDB) a basic autopilot and do it the old fashioned way. It is no fun from my perspective to anticipate and try to set up a complete route from dep, SID, ENR, STAR and Precision APP (RNAV or ILS) to find the kit does not cope, cannot load it or just simply cannot fly it and even it does, you cannot jump in a change anything. It is one of the reasons I have avoided the Airbus automation route with any Airbus aircraft in XPlane unless off course I am completely wrong but not sure how you can get install that Airbus logic over the top of a default FMS!

It is made worse by reason of the fact you cannot bolt it or addin a comprehensive substitute into the 3D cockpit or as a 2D pop up at least because of the need to get them all communicating and simply everybody seems to be hanging on Laminar doing it. Yes you can get a functional FMS-GPS unit on some models that is true but your also paying for the privilege. This is the only time I actually miss some aspects of xml coding and P3D stuff because their were some very clever and robust FMC-GPS units around (Like the Honeywell Unit or the CIVA INS). 

All said great work with the Boeing 747-400 can only get better I guess!

Edited June 9, 20233 yr by coastaldriver

Maybe these are good question to put to Austin for his upcoming interview...

I honestly am no longer on hold for any specific stuff... not even waiting... Time will tell and at most I will keep checking XP12 every time a worth testing update comes to live, and that's about all I do these days in XP.

Edited June 9, 20233 yr by jcomm

1 hour ago, jcomm said:

Maybe these are good question to put to Austin for his upcoming interview...

please, much more impact when such questions come from the community rather than just little old me.

1 hour ago, coastaldriver said:

Given that it so fundamental now to onboard flight navigation and aircraft management I find the lack of attention to this major Laminar cluster peculiar to say the least.

Its the nature of the beast tbh.

The current/next release (of the 744) is pretty much the first one I'm fundamentally satisfied with, almost everything I needed from Laminar to get to this point has been there from before I started on it, it's just taken a few years to actually leverage it. The #1 priority on Laminars side over that time has to get the sim to look the part, they are very nearly but not quite there yet, all the things they have been doing have been going in the right direction.

That's a huge change from the year or two before I started, when I raged a lot about how much effort they had put into getting oil barrels looking good - completely useless when X-Plane back then could barely manage 20fps and had mammoth stutter problems.

And they've been great, over that time several people have started and given up on a fuselage model rework, Laminar themselves are in the process of reworking that from scratch:

reworking stuff for the onboard navigation was not a priority when the sim is performing like #### and to quote @Greazer "looked like a PS2 title", Vulkan device losses on large numbers of systems and routes.

As those issues become ancient history it is natural and good that the biggest problems left are things no one ever really noticed before.

Edited June 9, 20233 yr by mSparks