Jump to content
Ray Proudfoot

Problem refining fuel burn

Recommended Posts

26 minutes ago, Ray Proudfoot said:

Even by ToC the difference is a few hundred pounds

 

At higher weights a 738 is going to consume somewhere between 1700 and 2000kg getting up to typical altitudes (and I've just been looking at a bunch of realworld OFPs to verify that). That matches the numbers in the performance file. You appear to have already added 5.6% fuel bias on the climb. If you are saying that you're burning a few hundred lbs more than that in the sim....

 

 I'd say CI 45 or 35 is high for normal planning purposes, but I'm open to correction. Every OFP I've been looking at is between 10 and 20. Having said that Climb at CI40 vs CI10 only adds about 3% to climb fuel burn.

Share this post


Link to post
Share on other sites

Gents,

Regarding that CI I can't recall where I got it from but it seems a reasonable value considering we don't actually pay for our fuel. In any case if it's 45 in PFPX and 45 in the aircraft then it's a like for like.

 

Take a look at the attached spreasheet and chart. I recorded the PFPX and actual fuel remaining for each waypoint on EGCC-EGPH. There is a difference when landing as I had to head west for a 06 landing which PFPX did not take into account.

 

Notice that as soon as I'm airborne the aircraft is eating more fuel that calculated. Between ToC and ToD the burn seems pretty accurate. I have an admission though. Climb was 250/280 but I forgot to change cruise to M0.79 so it was cruising at M0.72. I'm not sure how much extra the higher speed would consume.

 

The variation doesn't really change on descent so the problem is with the climb calculations. What do you think?

PFPX_Fuel.png

Share this post


Link to post
Share on other sites

From my own experience I can only suggest to look at probable causes Ray.

 

You are already applying a 5.6% bias whereas I have no alteration at all yet my figures are as close to a match as you could expect.

 

Make sure your physical throttle in not overriding the auto-throttle function, disable the manual override at least to test:

 

ngx-throttle.thumb.jpg.43c51a809193c68540aa9f25b18dc970.jpg

 

Check that the N1 is not exceeding that commanded by the auto-throttle in the climb.

 

 

Share this post


Link to post
Share on other sites

Hi Stephen, I’m just back home from a trip to friends so I’ll check those settings tomorrow and get back to you. The problem only applies to the climb phase which is something I’ve learned today.

 

No idea where that 5.6% bias has come from. How can I disable or remove it so it can be recovered if needed?

Share this post


Link to post
Share on other sites

Possibly from the APM use.

 

Use the PFPX Aircraft Editor and delete both 5.6% bias entries then save the aircraft.

 

PMDG do not recommend the use of FSUIPC for axis control, it may or may not work, it is certainly possible that its use could push the N1 beyond that commanded by auto-throttle. My own experience of that is varied so safest option is to use 'send to FS as normal axis' but experiment and observe the options yourself.

Share this post


Link to post
Share on other sites
18 minutes ago, srcooke said:

Possibly from the APM use.

 

Use the PFPX Aircraft Editor and delete both 5.6% bias entries then save the aircraft.

 

PMDG do not recommend the use of FSUIPC for axis control, it may or may not work, it is certainly possible that its use could push the N1 beyond that commanded by auto-throttle. My own experience of that is varied so safest option is to use 'send to FS as normal axis' but experiment and observe the options yourself.

 

Thanks for that. I do have everything calibrated via FSUIPC. Have done for ages. I did have a discussion with them about FSUIPC when I first bought the PMDG but the issues have all been resolved. Throttle wasn’t one of them. I have loads of controls mapped using their SDK via FSUIPC.

 

I’m pretty certain N1 does not exceed the limit. It’s probably that bias entry.

Share this post


Link to post
Share on other sites

Ray,

 

If you remove that bias, PFPX's calculated fuel is going to reduce - by 5.6% or typically somewhere around 100kg for the climb segment.

 

That is only going to increase the differential between PFPX's burn and your sim burn. At the moment given the way your sim is behaving you'd be wanting to increase the climb bias to somewhere around 10%. Something is not right, because your sim burn is high in all three flight phases, even in the descent you're burning 200 (units on the bar chart?) more than PFPX's calculation. I no longer have P3D installed, I now only use X-Plane, so I can't take a look I'm afraid.

 

Jon

Share this post


Link to post
Share on other sites
6 hours ago, srcooke said:

From my own experience I can only suggest to look at probable causes Ray.

 

You are already applying a 5.6% bias whereas I have no alteration at all yet my figures are as close to a match as you could expect.

 

Make sure your physical throttle in not overriding the auto-throttle function, disable the manual override at least to test:

 

Check that the N1 is not exceeding that commanded by the auto-throttle in the climb.

 

Stephen, I checked and all were the same as yours bar Set P3D Loc Crs which I had set to On.

 

1 hour ago, BW901 said:

Ray,

 

If you remove that bias, PFPX's calculated fuel is going to reduce - by 5.6% or typically somewhere around 100kg for the climb segment.

 

That is only going to increase the differential between PFPX's burn and your sim burn. At the moment given the way your sim is behaving you'd be wanting to increase the climb bias to somewhere around 10%. Something is not right, because your sim burn is high in all three flight phases, even in the descent you're burning 200 (units on the bar chart?) more than PFPX's calculation. I no longer have P3D installed, I now only use X-Plane, so I can't take a look I'm afraid.

 

Jon

 

Jon, that makes sense as I replanned the same route and it calculated 200lbs less fuel. I’ll find a way to amend the fuel burn to 10% and take it from there. What would you suggest for the drag %?

 

If you check that bar chart again you’ll see the difference didn’t significantly change after ToC. The bar chart is a rather crude tool with 100lb minimum on that axis.

 

I’m reasonably confident that with adjustments to fuel bias I can get it close enough. Why it differs from Stephen’s is a mystery. I believe the expression is “bloody computers”! 😁

Share this post


Link to post
Share on other sites

Ray,

 

I've never used drag bias - that may then also affect speeds. It is very hard to understand what's happening without seeing the whole picture. Fuel burn is only one element in the whole equation. The ideal would be if we had the OFP used as it is in the real world, as a How-gozit. With actual time and fuel, and some IAS/TAS/wind checks annotated onto a printed OFP it would be much easier to see exactly what's going on. However I appreciate that's something that only a few simmers would ever do.

 

Sorry if I'm repeating what Stephen has already asked and for doing the IT equivalent of "is it plugged in" but -

1. Does your PFPX weather match whats in the sim?

2. Does your takeoff weight match between PFPX and the sim - forget all the underlying fuel, weight, ZFW, etc, it's the TOW which is going to drive things. I just ran a 2 hour sector at a few different weights and the % increase in TOW results in a similar % increase in fuel burn.

3. PFPX and sim aircraft types match (winglet/non-winglet)

 

Otherwise based on your typical flight examples bump up the PFPX Climb fuel bias to around 10%, increase the cruise bias slightly, to maybe 7-8%, add an extra couple of hundred pounds for ATC getting you down low early on the approach and you should be there.

 

Jon

 

Share this post


Link to post
Share on other sites

Jon,

 

Stephen has asked for my G-VABA.txt file as well as asking questions. He may find something there.

 

PFPX Wx is using the weather service I pay for and I have AS for P3D v4. The departure and arrival weather match. I haven't checked for each waypoint.

 

I haven't checked TOW but basic empty weight does match. I'll check TOW this afternoon.

 

Yes, using a 737-800 winglet file for PFPX which matches the PMDG model.

 

It doesn't appear possible to add bias directly into the APM dialogue. Instead, various weights, speeds, temp and fuel burn have to be entered and the program works out the bias. Instead of entering all that data at cruise I'm going to enter it during climb as that's where I'm burning more than is being calculated.

 

More later.

Share this post


Link to post
Share on other sites

Matching Takeoff Weight is the bit that matters. The total weight of the aircraft is what drives the performance and fuel burn, irrespective of how that total weight is made up.

 

You add bias through the Aircraft Editor. Select the relevant aircraft in the Aircraft Database and then select Edit. Type the bias into the relevant data entry box, make sure you save when you have changed the appropriate value. APM has nothing to do with what we're proposing here, simply type in the suggested values. If you are finding that the TOW in your sim and in PFPX aren't matching then my suggestion is remove any bias. Do a flight with matching weight, see what percentage the burn in climb  and cruise differs and type that into the Aircraft Editor. Keep things simple.

 

One other thing just to be sure nothing external is messing with your PFPX setup -  did you use a template to create the aircraft? Using a Template to create an aircraft rather than creating straight from the performance file can sometime cause problems, personally I never use them, and I don't provide templates, only the .per files.

Share this post


Link to post
Share on other sites

Jon,

 

Looking at the Aircraft Editor dialogue there isn't a field for Climb Bias which is really what I need. Just Cruise Fuel Bias and Hold Fuel Bias. Confusing isn't it?

 

Weight is important as you say. I'm at EGCC for another flight up to EGPH. The PMDG CDU advises GW is 142.5. PFPX shows Ramp Weight is 142170 and if you subtract 2/3 of  Taxi fuel of 330 it becomes 220. 142170 - 220 = 141950 TOW. Not sure why PMDG is not showing a closer value although a few hundred pounds won't account for the problem.

 

I run Avliasoft's Electronic Flight Bag so can read instant weight values. I'll report back later. No Bias added. Just same as yesterday until I hear back from Stephen.

Share this post


Link to post
Share on other sites

A slight improvement on yesterday's flight. Still consuming too much fuel or PFPX not calculating suffucient.

 

TOW when holding on 05L was 141,900lbs. PFPX calculated 141,840lbs. I don't think 60lbs is going to make a difference so we can put that one to bed.

 

Just waiting for Stephen now who is using my aircraft txt file.

Share this post


Link to post
Share on other sites

Ray,

 

Planned with PFPX weather and flown ASP4 live at the scheduled time, NADP-1 as published.

 

There is a slight increase in fuel burn following the FMS implementing the STAR altitude restrictions, also I did not add a circuit in distance from MIQ which would have reduced the discrepancy on landing.

 

No issues with fuel burn on the climb out or cruise.

 

My data entries are highlighted.

EBBR-EDDM-2r.pdf

Share this post


Link to post
Share on other sites

Thanks Stephen. How frustrating. It does rather point to the aircraft I’m flying. But even if I switch to the house livery 738W it’s still using the same air file.

 

I think my only solution is to add a fuel bias and take it from there. Thanks everyone for all your help and advice.

Share this post


Link to post
Share on other sites

Thinking more about this fuel discrepancy problem adding more fuel isn't going to resolve it. The problem as far as I can see it is the amount of fuel that is being burnt in the climb phase. Thereafter, things seem okay.

 

I have TO-1 set and have tried both CLB-1 and CLB but they're not really making any difference. I imagine somewhere in the PMDG 737 files are entries relating to fuel burn. It would be interesting to compare mine to Stephen's. Maybe it's there the solution lies.

 

Thoughts?

Share this post


Link to post
Share on other sites

In my PMDG737-800W aircraft.cfg I have the following. Do others have the same fuel_flow_scalar setting?

 

[GeneralEngineData]
engine_type = 1
Engine.0 = -5.943, -16.470, -2.880
Engine.1 = -5.943,  16.470, -2.880
fuel_flow_scalar=1.2105
min_throttle_limit = -0.3719
max_contrail_temperature = -30

Share this post


Link to post
Share on other sites

I would not recommend changing the aircraft.cfg file Ray. According to PMDG that could cause all kind of problems with their aircraft as most of their calculations are done outside of FS.

My GeneralEngineData section looks the same as yours.


I assume you follow a normal climb profile, aka NADP, 250/M.whatever results from your CI, do you?

In my airline we always enter the TOC wind in the cruise wind field on the PERF INIT page to get a more economic climb speed resulting from the CI. Not sure if PMDG models the wind effects on the CI resultant speeds at all (I believe they do not), but this might make a small difference.

Then, if you get a reduced climb thrust (CLB1 or 2) you can clear that and use full climb thrust as soon as ATC clears you above FL150. From then on it's always full climb thrust, which saves a bit of fuel as well.

 

Do you retract the retractable landing lights when you raise the gear? The myth says these use an additional 50kg when they're extended until FL100. Whether that's true and whether PMDG modelled it is another question of course.

Share this post


Link to post
Share on other sites

Thanks Emanuel for that. It wasn't my intention to change those values. Just wanted to know if others have the same and from what you say they probably are.

 

Yes, normal climb profile of 250/300/.78 rather than 250/280/.78 but that is how the PMDG CDU is configured so I just go with that. CI is 45. I have the same set in PFPX and Topcat.

 

I also enter winds at ToC and the ISA DEV. All standard stuff really. I engage a/p within 30 secs of take-off and before that follow the FD.

 

I must confess to not controlling landing lights other than my single GoFlight T8 switch to turn them on/off. They go off around 8000-10000ft.

 

This is a most frustrating problem that we cannot resolve. I have P3D v4.4 with a standard install of the PMDG737-800W. I really don't know where else to turn except perhaps the PMDG forum. It doesn't appear to be a PFPX problem.

Share this post


Link to post
Share on other sites

Ray,

 

late in the discussion but do you if for pmdg there is a different .air file for the -800 an the -800w?

Share this post


Link to post
Share on other sites
16 hours ago, phil747fan said:

Ray,

 

late in the discussion but do you if for pmdg there is a different .air file for the -800 an the -800w?

 

Philippe, I believe I'm using the correct .air file but will check later today or tomorrow. For now, adding 500lbs of fuel is getting around the problem and I land with fuel remaining very close to that calculated by PFPX.

Share this post


Link to post
Share on other sites

Philippe,

 

I'm looking at the Aircraft Database in PFPX and the entry for the aircraft I'm using doesn't list an .air file. It simply includes weights etc which match those shown in P3D Vehicle selection.

Share this post


Link to post
Share on other sites

While it's not the ideal solution, have you considered simply to add a takeoff burn of those 500lbs into the PFPX profile of your aircraft?
PFPX has this option and then it'll generally add 500lbs to the fuel required for climb.

 

It may give you odd values when you fly short flights at lower cruise levels as you'll most likely burn less than those 500lbs, but for most flights it should work.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...