Hi, I followed iniBuilds tutorial on how to correctly start the descent from TOD but the airplane always arrives near the airport too high.
In the video the guy sets the target altitude and then presses PROFILE button near TOD in order to activate P.DES mode. I tried it, it doesn’t work.
Another possibility, for me this is the correct one, is to set the target altitude, pull the knob in order to arm the P.DES mode, and then the airplane automatically acitvates P.DES mode when reaches TOD. The beahviour is the same as above.
Here you’ll find a detailed description of what happens, I read from a real pilot that what all says about the bad VNAV performance of A310 is not real, the system is not perfect but in the real life it doesn’t act like in the simulation because in this way would never get certified.
So, for me there is a big bug but it seems that iniBuilds doesn’t take this into account.
Try to do this flight (30 minutes):
Fiumicino - Pisa, Cruise altitude 12500 feets
LIRF (34L) - LIRP (ILSY04R)
and you will arrive too high on the airport
I’ve done my own testing with regards to the VNAV on the A310. I’ve left it alone to do it’s thing and more often than not, I’ll end up being high to capture the glideslope. Whether this is a bug or not, I’m not really sure.
I’ve learned to arm the descent so the plane will begin the procedure on it’s own, but then I turn the profile off at about FL100. I adjust the vertical speed so I’m spot on to intercept the ILS.
Again, I’m not really sure if this is a bug, or just the way the plane is. I’m getting the impression that it’s not completely hands off the same as the A320 is?
IMHO this is a bug, I’m an aerospace engineer, not a pilot, but in this way the airplane is dangerous, I don’t believe that the real airplane has this big problem
Interesting! Yes, pilots fly the machines but engineers design them.
To my understanding from a post in a different thread the A310 behaves like its vintage realworld counterpart in the sense that the VNAV profile is okay for climb but not so much for descent. Keeping it on a VS of -2500 to -3000 works like a charm though.
To my understanding, the old software was rough enough to guide a blasting engine but less sophisticated to gently bring all that mass back down again.
If I (try to) use VNAV for the descent I end up with a TOD about 30 -40 miles before the airport. I usually do my descent calculations myself, do the typical rule of three with an ROD of 5*GS. It’s much preciser than these VNAV calcs
I understand the point of an old system/software but the fact that the real counterpart acts the same as the simulated one is “a voice” that iniBuilds said.
It’s impossibile that the real airplane brings you 5000 feets over the target, it’s extremely dangerous. All the systems must be certified and, if the real A310 works in this way, I can’t, as engineer, understand what was certified.
I found only one comment from a real pilot that confirms what I think. The other comments are all from simmer.
Anyway, could please someone of you do a flight with the flight plan of the opening post? That route puts the point on how bad the VNAV acts.
I’ve flown a couple of VNAV systems IRL and they can be very funny in how they act, airbuses in particular.
Rather than starting with, “Does it get me all set up for the ILS?” I’d ask, “Will it get me down a STAR without violating any restrictions?”
The difference is, in the 320 at least, the Airbus box is very cognizant of being in the Cruise phase, or the Descent phase, or the Approach phase. It would, theoretically, do a fine job were it all linked up from STAR transition all the way through the missed approach and allowed to slow when it wants to.
And yet, I have never been cleared in such a way that this was allowed to happen.
So in every arrival/approach I have ever done in the Bus, I have had to intervene with either speed or path, or both, to get the aircraft down.
So, I guess what I’m saying is, “What you think the 310 is trying to do, and what the 310 thinks it’s trying to do may not be the same thing.
Having said that, autopilots in most home sims I’ve ever played have always had horrendous latitude compared to what they do IRL. So, take all that for what it’s worth.
Read your linked post. Indeed it sounds like the Bus is “activating” the approach which will slow it to its “characteristic” speed; green dot if flaps are up, etc.
I don’t know how or why the MSFS box does this. In the 320, there is a point depicted on the ND that shows where the FMS will activate. Or you need to do it manually.
Also, you can enter selected (manual) speed and the FMS will still attempt to follow the path. This is what the overwhelming majority of pilots I fly with do. As I said earlier, the Bus sort of assumes that you will have the entire chain of transition, STAR, via, approach all linked up and it will be allowed to “do it’s thing”, which rarely happens.
Also also, what’s your descent profile showing? Is it showing you on or high on the path? The (any?) FMS won’t violate your speed to make path. It’ll take all the leeway it’s allowed and then just descend in what’s basically level change.
I looked at the arrival and the min altitude at the start of the STAR is 7500ft. So that’s 22NM by the rule of threes and 4-5NM to slow to start getting the flaps down. The STAR starts pretty close to 30NM. So doable, but not an abundance of slop in there. Do you get any “TOO STEEP PATH” or “DRAG REQUIRED” messages in there anywhere? In the 320, I think that they won’t come on unless more than half speedbrakes are deemed necessary.
So, if you are high on the path, the aircraft may just need half/full speedbrakes all the way down the chute. That’s not unheard of.
If, on the other hand, the Bus is showing you on path the entire way, then it believes it’s exactly where it needs to be and you are missing something in the programming.
Such as a procedure turn. If the box says that it is going to the VOR then proceeding outbound and looping back to the ILS, it may think it has all the time in the world to descend.
Using cost index of 25, like a simmer suggested, I correctly landed.
Before I’ve used 40
Hmmmm, odd. That shouldn’t alter your descent speed, certainly not that much.
But, you know the old saying:
“If it’s stupid, but it works…it’s not stupid.”
Well, glad it works as advertised!
Here the correlation, page 51:
Interesting, and logical.
In our company we use cost index for climb and cruise, but a fixed 290kts for descent (or as assigned/depicted), so we don’t really face this issue.
I didn’t read the whole source document, but I’m surprised that the TOD calculation doesn’t take this into account. In addition, and perhaps this is the range of CIs we typically use, I never see the descent speed difference vary more than 30Kts or so, top to bottom, before we change it that is. Considering the 320 has an allowed speed variation of +-20 (a 40kt window, max +5/-20 when facing a speed constraint), I still find it surprising that it’s that off between something like 25 and 40 CI.
It may be that the MSFS version is more sensitive to CI than the actual 310. And the 320 box that I fly is also at least a gen different from what I have seen of the 310, so who knows?
Anyway, glad you got it figured!
Allowing the A310 to automatically follow the descent profile does seem to result in being too high for the approach. Two changes I make to help are:
Change the flight plan height constraint for IAF or the waypoint at platform altitude from “At or above” to “At”.
Using spoilers to assist with any decelerations for speed constraints and once the approach has been activated. I watch the Engine gauges and deploy spoilers when the A/T has commanded idle and then put the spoilers away when the A/T is commanding thrust increase.
On recent A310 flights I’ve been using CI of 65 for the climb and most of the cruise as this gives the mach 0.8 that Simbrief uses. At this point fuel consumption is almost right on target. Before getting to the TOD though I change CI to 0 as I’d noticed that much more fuel is used on the descent than Simbrief predicts. I have noticed that the TOD is recalculated accordingly and comes earlier with a lower CI. Still uses more fuel on descent than predicted though but its closer.
It would be interested to know how real A310 counterpart works, maybe the CI is changed before TOD in order to have a better descend, who knows!
Is it possibile to find original A310 procedures?
Having tried various CI for descents I haven’t noticed any change in the aircraft’s ability to follow the descent. Changing the earliest approach platform altitude constraint to a hard “At” constraint has the biggest benefit. This makes sense as the A310 obeys the “At or above” but that often leads to it still being too high for the final approach. Using spoilers for the deceleration phases allows more time for descent although I don’t bother if the descent is looking ok.
That is an excellent suggestion and something we sometimes do as well to reduce the “latitude” that the FMS has to choose the descent profile. Some arrivals are notorious for getting you high.
I noticed earlier that those altitudes on the arrival are all at or above. So the aircraft is perfectly happy crossing them high. Making at least the last one a hard altitude would force the AP’s hand somewhat.
How can I change it to “At”?
In the 320, there are two ways.
- Type either a 3000 or /3000 into the scratchpad (assuming it was 3000 you wanted) and press the Line Select Key (LSK) on the right side of the line that contains the point in question. Some boxes allow you to truncate the last two zeros (/30 instead of /3000).
- Select the LSK to the right of the line that contains the point first. It will bring up a Vertical Revision page which will have several fields, one for an altitude constraint.
In either case, execute the change.
Of course, this will encourage the AP to cross the point AT rather than At or Above the altitude you entered. But it still will only go as far as going Idle and maintaining speed.
If that’s not enough, you’ll need speed brakes. But at least the VNAV path should be showing you accurate guidance to your entered altitude.