Once more helloI posted an example of an unflyable VNAV approach already a few days ago. Got no answer from PMDG!Therefore i decided to post Another example!I tried something simple: ILS34L in Seattle (KSEA). After takeoff up to 5000. and then via downwind to the approach transition to final. I only used LOC/VOR mode for lateral nav and VNAV for descend. All points and altitudes were calculated by FMC. And eventhough the V/S on DES page seems to be correct the VDEV bar on the ND was always ways to low!The preset runway elevation was correct an the altitudes gave me a FPA for the approach of about 3.4 degrees which would mean that I would fly above the G/S so why is the VDEV bar between 2000 and 4000 feet BELOW G/S?????????You want to know how the story ended? The plane struck the ground about 5 miles away from the airport and know what: THE VDEV BAR WAS PERFECTLY CENTERED, thanks god!!!!If you are all so sure that I am doing a mistake fly an approach for me and post some shots of the FMC and ND and ILS indication on PFD during that approach! They won't indicate the same!And before you tell me to use another mode, to check this or that value, RTFM or to try another solution show me if you can do it. I have a lot of FS experience and I know that I'm not wrong here! Alway worked in PIC767 always worked in DF737 even works in SSW A310 never woked in PMDG, why?ThanksMartin

Hmm. VNAV on finals? That's not right. VNAV should aim to get you down to the ILS capture alt, then you should switch to APP mode. I don't think the VNAV is meant to be used for final approaches, unless you are going in visual, in which case it should be switched to ALT HOLD at the MDA anyway and then discoed and hand flown onto the tarmac, regardless of what the A/P is doing you don't go under MDA until you can tell what color of eyes the TDZ has.As to why your VNAV was flying you into the ground 5 miles out I don't know. Can you post a view of the LEGs page? Or at least read off the relevent legs, with ALT/SPD restrictions for them.Also, what was your speed/flaps/gear config when you continued to drop to the ground? It might be you were slow and low, which the autopilot can't do much about, except to stall the plane.One or two screenshots on approaches at key points would be good.

Yeah, please post some screen shots of this strange approach procedure. I'm a mechanic by trade and my favorite motto is "if they can....they will". :-lol It's amazing the things people will try to do and then claim "it doesn't work". :-lol :-lol Willhttp://www.precisionmanuals.com/images/forum/ng_driver.jpg

"If you are all so sure that I am doing a mistake fly an approach for me and post some shots of the FMC and ND and ILS indication on PFD during that approach! They won't indicate the same!And before you tell me to use another mode, to check this or that value, RTFM or to try another solution show me if you can do it."From flight 1:http://forums.avsim.net/user_files/88257.jpgFrom flight 2:http://forums.avsim.net/user_files/88258.jpgAs recommended I use SPD INTV in these type of VNAV approches. At the time I took the shots it wasn't engaged.

As has been mentioned, VNAV will not fly a full precision approach such as an ILS. VNAV will not follow a glide slope. For a precision approach you need to use a different mode. For a non-precision approach VNAV can only take you to MDA. FYI: For an ILS you need to be using APP mode. VNAV and LNAV should only be used to position the aircraft to a point for an appropriate intercept of the LOC. Once positioned you should arm APP mode before intercept. APP mode will follow both the LOC and glide slope down for a CAT I,II or III landing. Or you can disconnect A/P at or before DH and hand fly her down. Also, your speed is a bit fast. In APP mode you need to use Speed mode and an appropriate IAS value for the approach. Pitch and power in this model are very important to ensure a smooth landing.From the screen shots above the product seems to be working perfectly.Finnally one suggestion, and I don't mean for this to start a flame war, just a friendly suggestion. :-)Before posting that the entire VNAV function is "Unflyable" it may be better to ask questions about how VNAV functions. Unless you are an NG pilot, you may not be the ultimate "Authority" on how VNAV is suppose to work and when it is used (or judging whether or not it is "Flyable"). Many of us have spent lots of hours flying VNAV successfully without any issues.

" And eventhough the V/S on DES page seems to be correct the VDEV bar on the ND was always ways to low!" Maybe it's partly because you have not RTFM throughly since in this case the V/S you see on the DEScent page only applies to the CURRENT NEXT WPT RESTRICTION unlike the VNAV DEV POINTER which is based upon the path calculated from TOD to the rwy. There is nothing wrong here and this is not 767 OR a 737 Classic, they ARE not alike although a updated Classic FMC is certainly more alike than a 767. 1] VNAV, MCP must be set to 0 PRIOR to the FAF. 2] If you have any step downs use VNAV SPD or use SPD INTV. 3] VNAV is NOT slaved to speed but to the PATH keep that very clear in your mind on final. How far are you tyring to take VNAV to? You cannot LAND in vanv so why don't you give us a step by step of what flight-route-PROCEDURE you are trying and then we can also see what is it you are trying to accomplish.Best Wishes,[h4]Randy J. Smith[/h4][h3]P M D G's 747-400[/h3][h4]coming to a runway near you[/h4][/font color]Caution! Not a real pilot, but do play one on TV ;-)AMD 64 3200+ | ASUS KV8 DELUXE | GFORCE 5700 ULTRA @535/1000 | Maxtor 6Y080M0 SATA 80 GIG | 512 DDR 400 | Windows Xp Pro | Windows Xp Pro 64 |

You should not that the general guidelines say that VNAV should only be used unless you have no margin of error. For an example, if VNAV would go crazy on your during your final approach, good luck on recovering. You should only be using VNAV for climb and descent. The final approach should be flown with V/S, or APP (for ILS)Nick B.Continental Airlines Virtual 737NG Pilothttp://www.precisionmanuals.com/images/forum/800driver.jpg--AMD Athlon XP 3200+ @ 2.2 Ghz (Equal to 2.8 ghz)400W Power supply3x 80 mm Case FansSoyo VIA KT600 Dragon PlusnVidia GeForce FX 5200 128 mb2 x 512 PC400100 GB Western DigitalMicrosoft Sidewinder Precision 2

Hi Martin,I think your question was answered already.But I saw you

Do you have a link or can you briefly explain how to execute an RNAV approach? I have seen RNAV charts for almost every airport I fly to/from.I guessed they were like GPS or V/LNAV approachs down to (what should be) glideslope at MDA. Then (it appears) you do a visual landing.Second semi-related Q, on the UK charts I can't seem to find the MDA, I can find lots of different classes for an OCH/OCA but I thought that was overhead circling height, for patterns if the runway changes or something. For MDA I was using the AlT for the MA on the charts.Thanks.PS: I flew from Belfast to Gatwick with VNAV/LNAV from 800ft on take off until I was over the runway threshold and VNAV discoed, then a visual land (all of pulling the throttles back and flaring). No problems at all with the approach (was a tiny bit off center line and a little fast VREF+11). I did get a few discos on descent, but that was a few dodgy ALT restrictions in the flight plan I loaded, which was an old one I had with a "decend in hold" at ASTRA.

Bj

Hi Paul,I think, Per

>You should only be using>VNAV for climb and descent. The final approach should be>flown with V/S, or APP (for ILS)The FAA disagrees ...LNav/VNav approaches are not new, says Jim Enias, FAA Flight Standards

>Also it's my understanding that these trad. techniques with>stepdowns, level-offs on non-precision approaches cause more>stress and workload for the crew, and may also be a cause to>more incidents. RNAV authorized approches in LNAV/VNAV may>reduce these risks. Maybe a rw driver can fill us in a bit>here?The few times i flew in an MCC ALSIM sim, i did this kind of approaches a number of times.I almost never really leveled up: the checks are so closely packed, that they are cleary meant to make you adopt a certain fixed rate of descent.Unless you really dive down a lot momentarily and then pull up, and then again; which is rather stressfull for the passengers, for the airframe, and for you.Also i wasflying in a general bi-turbine which is quite more responsive then a big commercial liner.So what i mean is that even if in the chart the vertical path is drawn as a series of stepdowns, i think that noone really fly them that way simply because it's not possible.At least this is my experience in a prof flightsim.

Hello againOkay, I hoped I would start such a discussion. I have flown the flight again, this time from Geneva to Zurich. With the same Approach on RW28 and this time, THAT S*** WORKED!!! :)I don't understand what I did different from my last approach but this time everything worked as it should.Just a few comments to the approach:- I know that the VNAV only helps you come down to the minimum. You'll see in my shots that this is what I intended to do.- The weather is real world with ActiveSky.- The plane as a ZFW of around 115000 lbs.- The FMC approach data comes from www.navdata.at- The approach chart is available under: http://swissfir.org/charts/LSZH/LSZH_VOR28.pdf- Speed and altitude changes are done as controlled by ATC in Zurich in real life.- flight deck handling is done as in real life (information from several 737 flight deck videos on approach to Zurich)Here are the pics:http://forums.avsim.net/user_files/88422.jpgStill on ground at LSGG I program the route.http://forums.avsim.net/user_files/88424.jpgOver EKRIT. The plane is not flying the VPATH even though I never changed something on MCP it is already down to 6000. But still no big problem!http://forums.avsim.net/user_files/88425.jpgThe plane is about to decelerate to 185 kts for the final turn later on.http://forums.avsim.net/user_files/88426.jpgSuccessfully established on final. FMC Speed, VOR/LOC, VNAV PATH are active. But still the plane is flying the rest of the approach down to the minimum always a bit to high (not bad but still not optimal).http://forums.avsim.net/user_files/88427.jpg400ft above the minimums the runway comes in sight. The plane is still about 290ft to high (also visible by 4 white papi lights). Autopilot was then disconnected and landing was flown manually. (AND SUCCESSFULLY!)The approach was not bad that time. When I departed in Geneva I hoped to show you all how bad the plane behaves. I was surprised myself how well it worked this time. But I did exactly the same approach three times before without any success!Even in this approach the autopilot is not perfectly maintaining the vertical path, especially on short final he was about 300 feet to high. If the cloud base would have been lower the a landing with such an altitude difference is already difficult.On the other tries the plane always came in ways to low. I don't know why not this time but still the VNAV is UNPRECISE!What do you all think?Thanks and enjoy the shotsMartin

Hi Claudio, and thanks for the input.Yeah, I know that in regular operations you will adopt a fixed rate of descend and will not level off at each step-down. This is also how this apply to our sim. What I meant to say was that level segments may cause more workload.I also had in mind the fact (as stated in the above document) that the RNav concept is expected to provide more accurate aircraft position data and more stable descent paths than instrument approaches using traditional ground-based navaids.Some words from the Bulfer profile:Using V/S or LVL CHG-Disadvantages:-Each stepdown must be entered in the MCP ALT select.-Tell your teammate to remind you to reset the MCP ALT after each ALT HOLD.-Monitor stepdowns using the green altitude trend vector.-Though you can use V/S to follow the PDI inside the FAF, I'd recommend transitioning to VNAV PATH at the FAF to maintain the constant vertical angle to the threshold. This will lower your workload.Flying a constant angle vertical path using VNAV is an excellent tool. There's no need for a VDP and there is no level-off segment."BTW, this really turned out to be an interesting thread, he? :-)