Jump to content

Concorde in P3D v3.4 - pitching up on take-off roll


Ray Proudfoot

Recommended Posts

Ray Proudfoot

Really strange problem with Concorde in P3D v3.4. As I start my take-off roll the nose starts to come up even before 80kts. Unless I push forward quite forcibly on the yoke it will over-rotate and end up vertical by v1.

This is a short subsonic hop from EGLL to EIDW but I've flown this many times in FSX without problem. Everything is correctly trimmed as far as I can see. Anyone else had this issue? In all other respects the flight was fine.

Link to comment
Fraser Gale

Yes, there is something wrong with the physics at light weight as the rotation begins on its own, even although the delta wing produces 0 lift until it is presented to the airflow at a particular angle of incidence - around 10 degrees and above at takeoff speeds. Nothing should happen until the elevons are raised.

I have had this issue for a couple of versions though, not just 3.4

Frazz

Link to comment
Ray Proudfoot

Thanks Frazz for confirming this problem. But it's such a severe one I don't understand why many others have not reported it. And why only in P3D? I was on the beta team and flew extensively but these were long flights with a lot of fuel and I was flying FSX of course.

It appears this problem may have been missed but given its severity I really do think it needs addressing. Having Concorde sitting on its tail at 130kts is not my idea of realism. :(

How much fuel do you need to load before the problem subsides?

Link to comment
AdrianSmith

Like Ray I have never noticed it in FSX, sure the nose lifts marginally because all the thrust is below the drag line, but all the wheels stay on the ground until I decide otherwise.

What was your CofG?

Link to comment
Ray Proudfoot
7 hours ago, MrNuke said:

It isn't limited to P3D...

I have a test flight from EGLL to EIDW that I have flown many times in FSX and not had this problem. Maybe, just maybe the CG is the reason. In FSX it was 52.5 and there was no problem whereas in P3D it's 53. But 53 is the correct CG for that TOW according to BA rules.

I'll try it again in P3D with 52.5 and in FSX with 53 and see what happens.

MrNuke, if you've experienced it please give more details including TOW, fuel load and take-off CG.

Link to comment
Ray Proudfoot
7 minutes ago, AdrianSmith said:

Like Ray I have never noticed it in FSX, sure the nose lifts marginally because all the thrust is below the drag line, but all the wheels stay on the ground until I decide otherwise.

What was your CofG?

CG was 52.5 if you're asking me Adrian. For anyone who has CPS or CPS-X and wishes to try this test flight here is my plan:-

EGLL CPT UL9 KENET UN14 OKTAD DCT MEDOG UL18 LANON DCT ABLIN DCT LAPMO EIDW

Link to comment

Yep, confirmed, there is a pitch-up during acceleration in P3D at light weights. However, this happens if the CG is rather in the aft position. Moving it more to the middle prevents the pitch. I think that the pitch-up is not attributed to the wings producing lift too early, but to the aspect that the engines are so powerful that the back of the airplane accelerates faster than the front - much like a bike powered by the rear wheel that accelerates very fast (so that it pitches up, too). This is not necessarily realistic for Concorde, but as I said, moving the CG frontwards can help avoiding the pitch.

Link to comment
Ray Proudfoot
12 minutes ago, afterburner said:

Yep, confirmed, there is a pitch-up during acceleration in P3D at light weights. However, this happens if the CG is rather in the aft position. Moving it more to the middle prevents the pitch. I think that the pitch-up is not attributed to the wings producing lift too early, but to the aspect that the engines are so powerful that the back of the airplane accelerates faster than the front - much like a bike powered by the rear wheel that accelerates very fast (so that it pitches up, too). This is not necessarily realistic for Concorde, but as I said, moving the CG frontwards can help avoiding the pitch.

afterburner, whilst I accept that moving the CG forwards may lessen  or eliminate the problem a CG of 52.5 was not used by BA. 53, 53.5 and 54 were used.

I suppose as a stop-gap measure I will have to use 52.5 to stop Concorde pitching up so much that it's near vertical by Vr. The difference is immense. I just hope that this can be tweaked by Andrew.

Link to comment
Ray Proudfoot

A question to those who have experienced this pitch-up problem. Have you raised a ticket with the support team? If so, what was the response? If not, why not?

I mean no disrespect with that last question but given the accuracy of the flight model a problem as serious as this does need bringing to their attention. If no-one has I'm happy to raise one.

 

Link to comment
Ray Proudfoot

Just tested this scenario in FSX:SP2. With a CG of 52.5 the take-off roll is fine and there is no tendency for the aircraft to pitch up prematurely.

However with a CG of 53 it's an entirely different scenario. With no input on the yoke from me this is how the aircraft looks as it leaves the ground.

It seems some tweaking is required and hopefully soon. In the meantime I'll have to keep CG to 52.5 for light loads.

Concorde_CG53.png?dl=0

Link to comment
Ray Proudfoot

I tried the same CG of 52.5 in P3D and the nose still starts to come up before full engine power has been realised. Not as bad as CG of 53 but still disappointing. To get around this I'm going to have to push the yoke forward to keep her grounded until Vr. :(

Link to comment

I may have found the cause of the pitch-up: If you open the aircraft.cfg-file of Concorde-X and scroll down, you will find the following entries:

Quote

[GeneralEngineData]
engine_type=1
Engine.0=-7.000, -45.000, -20  // Was -20.000, -3.800 
Engine.1=-7.000, -41.000, -20  // Was -16.000, -3.800
Engine.2=-7.000, 41.000, -20   // Was 16.000, -3.800
Engine.3=-7.000, 45.000, -20   // Was 20.000, -3.800
fuel_flow_scalar=1.200000 
min_throttle_limit=-0.460000
max_contrail_temperature = -100

The "Engine.X=..." entries specify the location of each engine (longitudinal, lateral and vertical coordinates) relative to the datum reference point (refer to the P3D SDK here for more information). The values are in unit feet. To me, it looks like the engines are placed too far below the fuselage centerline (20 feet) and too far apart from each other (82 ft inner engines, 90 ft outer engines). This doesn't correspond to the real coordinates; I don't know the exact values, but looking at this sketch, the distance between the inner engines should not be higher than 30 feet (or 15 feet from the center), and there is no way that the height of the engines is 20 feet below the fuselage centerline. If you hypothetically place the engines of Concorde so far down and apply full-thrust afterburner power at light weight, it is no wonder that the engines will generate a torque that will pitch up the aircraft immediately.

If you look at the entries, you can see some values behind the double slash ("...// Was -XX.XXX, -X.XXX"). I am not sure, but this indicates to me that these values were used at some point in the past, but have been commented out and replaced by other values since then. The old coordinates actually correspond much more precisely to the real-world values. And indeed, if I replace the actual coordinates by the old, commented-out coordinates behind "Was..." (referring to lateral and vertical position), I experience no pitch-up at light takeoff-weights even at 54% CG.

I don't want to speculate why the old values have been replaced - only the developers can tell us the motive behind using these coordinates (this is why I don't recommend to replace the values on your own until the matter is clarified. If you decide to do so, do it at your own risk).

  • Like 2
Link to comment
Ray Proudfoot

Thanks for that analysis afterburner. I'm guessing the data is what it is to get the best of of the aircraft in all stages of the flight. I've never had a problem with any of it until CG was moved further forward on light loads. Perhaps the team felt it was best to configure Concorde for longer flights when she would be heavier and the short hops with light loads were compromised.

Would be nice if the developers could comment.

  • Like 1
Link to comment
Fraser Gale

Sorry I haven't gotten back to you sooner, busy here. 

Pit sounds like afterburner has found the problem but I think it does need a ticket opening to get it fixed by the development team. We really shouldn't be having to play around with the cfg files. 

I always used a CG of 53% as it is the legal forward value and I always got pitch up below 130T or so. I started using an abnormally low trim setting (3.5 down) to counter this and I don't know why I didn't report it. I'm more in the business of flying transatlantic like she was built for so I didn't have to put up with it too much. 

Incidentally, what won't help is that for 53% take-offs the trim is set at 1.5 down rather than 2.5.

Frazz

Link to comment
Ray Proudfoot

I agree with you Frazz that this is too serious a problem just left to discuss here. I will raise a ticket shortly.

The other problem I've found is the speed just gets away from you after rotation and soon hits 300kts even with the nose pitched up to 18 degrees or more. Ouch! :(

These two major issues make this Concorde more like the one Microsoft chucked into FS2000 rather than the finely honed version it is on longer flights.

If I told you the elevon trim was -0.5 you will probably shudder but that is the correct setting for this weight. No wonder she was so eager to get airborne.

Link to comment
Ray Proudfoot

Afterburner,

How do those engine numbers in P3D compare to the FSX ones assuming you have that version?

If you don't I'll fire up my computer and have a gander.

Link to comment

You mean the fuel flow numbers or the engine coordinates?

[Addendum: I guess you are talking about the engine coordinates. They are identical in P3D and FSX].

As far as the speed problem, do you apply the noise abatement procedure after takeoff?  (The Concorde Performance Calculator recommends to switch off reheat 48 seconds after start and lower N1 to 90% if you take off from EGLL at a takeoff weight of 125 T). If you take off at full power + reheat, you will quickly reach 300 knots - and this is not abnormal.

Link to comment
Ray Proudfoot
26 minutes ago, afterburner said:

You mean the fuel flow numbers or the engine coordinates?

The engine coordinates. I've started my PC and here are the numbers in FSX's version.

[GeneralEngineData]
engine_type=1
Engine.0=-7.000, -45.000, -20  // Was -20.000, -3.800
Engine.1=-7.000, -41.000, -20  // Was -16.000, -3.800
Engine.2=-7.000, 41.000, -20   // Was 16.000, -3.800
Engine.3=-7.000, 45.000, -20   // Was 20.000, -3.800
fuel_flow_scalar=1.200000
min_throttle_limit=-0.460000
max_contrail_temperature = -100

As far as I can see they're exactly the same.

Link to comment
Fraser Gale
6 hours ago, Ray Proudfoot said:

I agree with you Frazz that this is too serious a problem just left to discuss here. I will raise a ticket shortly.

The other problem I've found is the speed just gets away from you after rotation and soon hits 300kts even with the nose pitched up to 18 degrees or more. Ouch! :(

These two major issues make this Concorde more like the one Microsoft chucked into FS2000 rather than the finely honed version it is on longer flights.

If I told you the elevon trim was -0.5 you will probably shudder but that is the correct setting for this weight. No wonder she was so eager to get airborne.

Ray, I'm afraid I have to disagree with your trim setting, it was not a function of weight but a function of the takeoff CG.

Three options for takeoff:

53% - 1.5 degrees down

53.5% - 2.5 degrees down

54% - 2.5 degrees down

Those were the only options in BA as they gave stable characteristics for nailing the theta 2 target. If AF did it differently then so be it but I think the thought in BA was the less options the less can go wrong. 

Generally at light weights the general noise abatement procedure would be used rather than a noise time, meaning you would rarely reach 20 degrees nose up by the power cut at 1000ft. If you were going straight to climb speed you could cut reheat at 500ft and continue to accelerate.

Frazz

  • Like 1
Link to comment
Ray Proudfoot

Thanks Frazz. I think the trim setting may need looking at by @Pierre Chassang. I'm not sure how he calculates it but I'm sure he can check his AF rules in case they differ from BA's.

Point taken about the noise abatement procedure. Do you happen to know the TOW that triggered altitude rather than time? If we can get these rules clarified it will certainly help.

Link to comment
Seth Goodwin
12 hours ago, Ray Proudfoot said:

If I told you the elevon trim was -0.5 you will probably shudder but that is the correct setting for this weight. No wonder she was so eager to get airborne.

As Frazz noted trim setting was set with takeoff CG not weight*. Change the takeoff form in CPSX to AF and it becomes quite clear.

cgaf.JPG.dc3b30ef922c4ae852a7a45248e5fe4c.JPG

That said, I do wonder if the higher trim values with the lower cg's are part of the issue in the modeling. I'll do some test flights tomorrow on FSX:SE.

5 hours ago, Fraser Gale said:

Three options for takeoff:

53% - 1.5 degrees down

53.5% - 2.5 degrees down

54% - 2.5 degrees down

Those were the only options in BA as they gave stable characteristics for nailing the theta 2 target. If AF did it differently then so be it but I think the thought in BA was the less options the less can go wrong.

AF just apparently had double the options. I don't know how often the 3 unique values were actually used. This site has 3 examples and all 3 use either 53, 53.5, and 54. And as seen above the 3 common values between the two carriers have identical trim settings.

5 hours ago, Ray Proudfoot said:

Thanks Frazz. I think the trim setting may need looking at by @Pierre Chassang. I'm not sure how he calculates it but I'm sure he can check his AF rules in case they differ from BA's.

I'm not really sure there is anything for Pierre to check out on the trim settings. On the fuel page it looks like he is taking the block % mac and rounding up to the nearest takeoff % mac. He also obviously lets you manually select a TO % Mac or takeoff CG on that page as well. If you want to constrain yourself to the 3 BA options it is already very easy to do so.

 

The more interesting information that Pierre may have access to is when/if AF ever used the 3 other settings. On my link above for a relatively light 131,000 kg takeoff on a CDG-CDG boucle charter, they used 53.0%. Hypothetically, if they were unused or even abnormal it may make sense to eliminate them from the program especially if the lower CG settings result in takeoff trim issues.

 

*Weight does factor into it though as under 140,000 kg was a 53.0% CG takeoff for BA, high fuel + high load= 54%  and anything in between (based on a table) was 53.5%.

Link to comment
Fraser Gale
6 hours ago, Ray Proudfoot said:

Thanks Frazz. I think the trim setting may need looking at by @Pierre Chassang. I'm not sure how he calculates it but I'm sure he can check his AF rules in case they differ from BA's.

Point taken about the noise abatement procedure. Do you happen to know the TOW that triggered altitude rather than time? If we can get these rules clarified it will certainly help.

They do differ, I was already aware that the AF crews had more CGs to select from although I'm not sure what parameters they used to do it. In my personal opinion, I think it is always safer to have fewer options as it limits the chance of error, so I'll stick with the BA rules.

There aren't hard and fast rules that I have seen for deciding on noise procedure but I think it was generally accepted that if you got to 5 or more tonnes below the lowest weight in the specific noise charts, you would use the general procedure. If I remember correctly, the lowest weight for 27L at LHR was 120T so if you went below that you could change to the general procedure. 

The other thing I'm wondering about, is that when performing the takeoff calculations for weights below 140T there is a factor by which the RTOW less than PLTOW is changed to recalculate all the speeds. This is because the speeds are not accurate for lighter weights unless this extra weight factor is deducted. I assume that CPS calculates this for you? 

Frazz

Link to comment
Ray Proudfoot

@MrNuke, My EGLL-EIDW flight's Elevon setting was -0.5 for a TOW of 115.6T. I'll speak to Pierre privately about this. It's wouldn't be appropriate here.

@Fraser Gale, thanks for the info. Regarding your last sentence I'm not sure. Maybe Pierre can answer that.

Link to comment
Pierre Chassang
3 hours ago, Fraser Gale said:

...
The other thing I'm wondering about, is that when performing the takeoff calculations for weights below 140T there is a factor by which the RTOW less than PLTOW is changed to recalculate all the speeds. This is because the speeds are not accurate for lighter weights unless this extra weight factor is deducted. I assume that CPS calculates this for you? 

Frazz

Hi Captains,

Do you have documents concerning that point.

52 minutes ago, Ray Proudfoot said:

@MrNuke, My EGLL-EIDW flight's Elevon setting was -0.5 for a TOW of 115.6T. I'll speak to Pierre privately about this. It's wouldn't be appropriate here.
...

The actual CPS-X version has a bug on this.
The trim is still calculated with the 15 TOCG value list.( As for AF).
I immediately fix this.

Cheers

Pierre

  • Like 1
Link to comment
Fraser Gale
7 hours ago, Pierre Chassang said:

Hi Captains,

Do you have documents concerning that point.

The actual CPS-X version has a bug on this.
The trim is still calculated with the 15 TOCG value list.( As for AF).
I immediately fix this.

Cheers

Pierre

I do, the real BA take-off performance form was (at least in the early 1990s) a two sided A4 sheet which lead you through the calculations. The rear page gave instructions for departures at light weight. I don't have time to upload a copy at the moment but will as soon as I can. 

Frazz

  • Like 1
Link to comment
Fraser Gale

@Pierre Chassang Here is a link to the full take-off performance form, sorry but my PDF printer changed the margins and chopped it up for some reason but the information is there.

The crew would normally fold the paper so that only the main bit was visible for departure, a bit like the one on CPS looks like above.

https://www.dropbox.com/s/vzl2ol5n6hywf5b/Airbus LPC140815 copy.pdf?dl=0

Frazz

  • Like 1
Link to comment
Pierre Chassang

Investigating into my Air France doc, I found an equivalent rule but expressed a little bit differently.
indeed Air France applied the same correction but when the TOCG was 52.5 to 53.4%.
As those TOCG was linked to TOW< 140000 kg the result is the same.

Capture.thumb.JPG.1c27b0cb184023473c61350c5c8f0eff.JPG

The main difference is the TO speeds were not modified with Air France while they were with British Airways (at least for V1, Vr and Theta2).

I will implement this in the next CPS-X update.
 

Kind regards
 

Link to comment
Fraser Gale

Seems strange that AF didn't use the new PLTOW to calculate the new speeds, a couple of knots difference can save wear on tyres and improve RTO performance. Anyway, as I've said before, the two airlines did things very differently with certain things.

I'm glad I have been able again to contribute to your programme in order to make it even more accurate. 

Makes you wonder what else may be lurking...

Frazz

Link to comment
Steve Prowse
On 19/04/2017 at 2:01 AM, Ray Proudfoot said:

I've raised a support ticket for this problem. Hopefully it will be addressed soon.

Hi Ray,

Have you had any response from the developers yet?  How are you getting along with P3D?

Steve

Link to comment
Ray Proudfoot
59 minutes ago, Steve Prowse said:

Hi Ray,

Have you had any response from the developers yet?  How are you getting along with P3D?

Steve

Hi Steve,

No response whatsoever from the support team yet. Bit disappointing.

I've had a lot of scenery to install so have flown very little. But fps is good and it's noticeably smoother than FSX. I'm 85-90% done so should be flying by early next week or maybe over the weekend.

Link to comment
Steve Prowse
17 hours ago, Ray Proudfoot said:

Hi Steve,

No response whatsoever from the support team yet. Bit disappointing.

 

It is very disappointing Ray, but I don't think the 'team' will update our Concorde again.  At best they will update the installer as when necessary to suit P3D.  The rest is now very much bought as seen. 

17 hours ago, Ray Proudfoot said:

Hi Steve,

I've had a lot of scenery to install so have flown very little. But fps is good and it's noticeably smoother than FSX. I'm 85-90% done so should be flying by early next week or maybe over the weekend.

Best of luck with that, I'm very much looking forward to see how you rate P3D when flying Concorde.....Seeing so many guys having issues with Concorde on P3D reminds me of flight 1's ATR72-500 which was designed for FS9.  It was in its day the best Turbo I owned it was just great in FS9.  When FSX came on the market there was a rush to port it across, it was never the same  in FSX.  The same thing is true for our Concorde it works well in FSX, at least for me it does,  it will always have problems in P3D, it simply wasn't designed for P3D, so it will always have issues, I think..  So yes I'm waiting to see what you think of it in terms of actual flight/performance  not 'it looks so much better' etc;  you know the normal comments from the P3D guys. 

All the best

Steve

Link to comment
Ray Proudfoot
6 hours ago, Steve Prowse said:

It is very disappointing Ray, but I don't think the 'team' will update our Concorde again.  At best they will update the installer as when necessary to suit P3D.  The rest is now very much bought as seen. 

Best of luck with that, I'm very much looking forward to see how you rate P3D when flying Concorde.....Seeing so many guys having issues with Concorde on P3D reminds me of flight 1's ATR72-500 which was designed for FS9.  It was in its day the best Turbo I owned it was just great in FS9.  When FSX came on the market there was a rush to port it across, it was never the same  in FSX.  The same thing is true for our Concorde it works well in FSX, at least for me it does,  it will always have problems in P3D, it simply wasn't designed for P3D, so it will always have issues, I think..  So yes I'm waiting to see what you think of it in terms of actual flight/performance  not 'it looks so much better' etc;  you know the normal comments from the P3D guys. 

All the best

Steve

Hi Steve,

I spoke to Andrew at the Cosford show last October and he assured me they plan on a final bug fix. Knowing the person he is I don't think he would say that unless he meant it. Circumstances may change but if they weren't going to update it again I think you'd see an announcement. In any case P3D 64-bit is in beta so I'm sure they would want to make Concorde compatible with that as it would solve all those OOMs when running at 4K resolutions. I've given them a nudge on my support ticket.

The installation and configuring of all my scenery has made it a horrendous job. Think of 3 years of scenery purchases compressed into two weeks. I was low on SSD space too so needed to uninstall before installing again and getting all the product codes slowed things down but I'm 90% done now.

Only had a chance of one flight - KSAN to KSEA. Some things are different and it's perplexing why. For example I can't seem to get the same forward view as in FSX so I can see the nose. I use SimpleCam to configure and save views and in theory it should be identical to FSX but it's not. Not a major issue but one I need to fix. In terms of stability P3D is wonderful. Not had a single 'crash' yet and I've not run out of VAS either although it was getting low as I approached Seattle. But over 500Mb on landing so that's fine.

The flight model seems more powerful than that in FSX. On lighter fuel loads you really have to pitch up and even then it's difficult to keep below 250kts. I should probably test this on the EGLL-KJFK run as that may give me a better feel. But it has been tested in P3D and according to others it's the best platform, not FSX.

The cockpit lighting is nice. I've turned off HDR as everything was gloomy. There is a utility called PTA that you can use to tweak lighting. I'll look at that in slower time. Performance is great with unlimited selected. Seems happier than FSX locked at 30. of course the new Nvidia 1050 8Gb card is taking a fair bit of work off the CPU which is the main reason I switched to P3D.

So far so good. :)

Link to comment
Steve Prowse
16 hours ago, Ray Proudfoot said:

Hi Steve,

I spoke to Andrew at the Cosford show last October and he assured me they plan on a final bug fix. Knowing the person he is I don't think he would say that unless he meant it. Circumstances may change but if they weren't going to update it again I think you'd see an announcement. In any case P3D 64-bit is in beta so I'm sure they would want to make Concorde compatible with that as it would solve all those OOMs when running at 4K resolutions. I've given them a nudge on my support ticket.

 

I really  hope you're right and I'm wrong Ray. 

 

16 hours ago, Ray Proudfoot said:

 

The flight model seems more powerful than that in FSX. On lighter fuel loads you really have to pitch up and even then it's difficult to keep below 250kts. I should probably test this on the EGLL-KJFK run as that may give me a better feel. But it has been tested in P3D and according to others it's the best platform, not FSX.

The cockpit lighting is nice. I've turned off HDR as everything was gloomy. There is a utility called PTA that you can use to tweak lighting. I'll look at that in slower time. Performance is great with unlimited selected. Seems happier than FSX locked at 30. of course the new Nvidia 1050 8Gb card is taking a fair bit of work off the CPU which is the main reason I switched to P3D.

So far so good. :)

 

Thanks for the update Ray, I'll be interested to see how well you think it flies after a few flights.  Oh, hope you've got all your scenery installed now, what a nightmare you must have had..

All the best

Steve

Link to comment
Steve Prowse

So Ray, how are you finding P3D and Concorde?  Is it a marriage made in heaven, or are you repenting at leisure?  You know that the only Jet I use on FSX is Concorde and I also fly the Dash Q4 (majestic), plus some GA.  So for someone like me is the change worth £150+-?  Oh and any news from the Concorde development team?

All the best

Steve.

Link to comment
Ray Proudfoot

Hi Steve,

The weather has been so nice this week I haven't flown her. But apart from two small issues I'm entirely satisfied with her in P3D. Those issues are:-

1. Excessive pitch-up on take-off roll when lightly loaded. I never recall having that problem with FSX. That support ticket remains unanswered.

2. This sounds strange but I'm struggling with the Captain's View to get a satisfactory viewing angle for landing using Avliasoft's SimpleCam . It's almost like P3D has a different default zoom setting to FSX. It does make landing slightly more difficult.

The operation of the aircraft seem identical to FSX. The main benefit of switching to P3D is the handling of VAS and the extra work done by the graphics card. Performance is smoother. Those alone makes the move worthwhile.

To be honest the number of lightly loaded flights I will have that rotation problem is not a major issue. But it should at least be answered when an official support ticket has been raised.

Link to comment
×
×
  • Create New...