PMDG 747 Fuel calculations

[edgarvsfreitas 0]

Hello everyone.

 

 

I have a question about the 747. I'm a 747 addict and I fly the PMDG 747 all the time. It's the only aircraft I fly, so I'm trying to build an excel spreadsheet to calculate fuel for this aircraft. I have built my own excel sheet but I have errors of up to 10000lbs -sometimes less- in the total amount of fuel needed for the flight. What I would like to ask you is, how do you calculate the fuel for this aircraft? Is it exaclty by the manuals? All my calculations are based on the PMDG 747 manuals, but since it gives me this large error, I don't know if I'm doing something wrong. I've found a fuel calculator from Felice Valente and it seems very accurate, but comparing with mine, I have a big error in relation to that one.

I've attached a picture of Felice Valente's fuel calculator and another picture of mine.

 

My spreadsheet: https://drive.google.com/open?id=0B6JMHUkV2NKJejJJVzF0YTRhZzg

 

Felice Valente calculator: https://drive.google.com/open?id=0B6JMHUkV2NKJQ3N5YXhmWDMwV28

 

 

I was always getting an error when posting images here, that's why I posted those links instead of the actual picture.

 

 

Thank you very much in advance.

 

 

Kind regards,

Edgar Freitas

 

 

Ps: If you need any information from me, feel free to ask.

 

 

[skelsey 765]

As a rule of thumb, 12 tonnes per hour works well (quite a bit more than the real thing).

 

It's quite difficult to create an accurate spreadsheet etc for the PMDG 747 as it burns a bit more than the real thing, but not a consistent amount. I found a bias figure of about +8-9% worked well in PFPX and Simbrief.

 

All the above for the RR engines.

[downscc 1,327]

I did the same thing almost 10 yrs ago, it is still in the AVSIM library look for either Dan Downs or download filename 744fuelplannerv1.zip, I managed to get within about 3% using linear regression.  I haven't used it in years and could probably not answer too many questions about it but you can use/modify as you wish.

[edgarvsfreitas 0]

As a rule of thumb, 12 tonnes per hour works well (quite a bit more than the real thing).

 

It's quite difficult to create an accurate spreadsheet etc for the PMDG 747 as it burns a bit more than the real thing, but not a consistent amount. I found a bias figure of about +8-9% worked well in PFPX and Simbrief.

 

All the above for the RR engines.

Dear Simon,

 

Thank you very much for replying. If the calculation method present on the PMDG 747 manuals are correct, I think it's a good way to determine the total fuel load. I've managed to do the spreadsheet but my goal was to decrease the error I'm having.  Yesterday I've been working on the spreadsheet again and the error now is of around 4000lbs so I think it's not that bad. 

 

My goal is to have a fuel calculator as precise as it can be. 

 

Again,  thank you so much for the advice you gave me. 

 

Kindly, 

 

Edgar Freitas 

I did the same thing almost 10 yrs ago, it is still in the AVSIM library look for either Dan Downs or download filename 744fuelplannerv1.zip, I managed to get within about 3% using linear regression.  I haven't used it in years and could probably not answer too many questions about it but you can use/modify as you wish.

Hello Dan,

 

Thank you for taking some time to reply to my question. 

Actually I had already downloaded your fuel calculator before but only yesterday I managed to look at it more closely than before.  As I added the tables from pmdg to excel,  I've managed to make it give me the all the information I need by adding just elements like distance,  wind information,  FL And contingency time. I've compared your sheet with mine and the results are similar.  I'll make my sheet give me a bit more of fuel in the trip fuel just to be sure. ;-) 

 

Anyway, I appreciate your kindness by providing your fuel planner. It was surely a great help.

 

Kind regards, 

 

Edgar Freitas 

[WarDog 6]

As a rule of thumb, 12 tonnes per hour works well (quite a bit more than the real thing).

 

It's quite difficult to create an accurate spreadsheet etc for the PMDG 747 as it burns a bit more than the real thing, but not a consistent amount. I found a bias figure of about +8-9% worked well in PFPX and Simbrief.

 

All the above for the RR engines.

 

8-9% is far from "not a consistent amount",  1% on a 100T flight is 1T.

[edgarvsfreitas 0]

So what do you suggest?

[skelsey 765]

 

 

8-9% is far from "not a consistent amount",  1% on a 100T flight is 1T.

 

Sorry, I didn't make myself very clear.

 

What I meant was that about +8% (in Simbrief) and about +9% (in PFPX) worked well for me in terms of giving me a sensible overall fuel load.

 

However, the problem with creating a truly accurate profile for the PMDG 747 (if you're basing it off real-world data) is that it is not as simple as saying that "it burns 8% more than the book says", because the amount it burns more than it should varies over the course of a flight. So 8% (or 9%) is an average figure which will get you in the ballpark for a typical stage length of about 8hrs.

 

However, from what I recall the PMDG 747 burns far too much fuel (maybe book +10% or +12%) in the early stages of the flight. As the flight progresses, the fuel burn (as expected) decreases, but not at the same rate as the fuel burn decreases in the real thing. So after 4 hours, perhaps, it might be burning book fuel +8%. After 8 hours the fuel burn will have decreased further, to perhaps book + 4%, and so on.

 

So if you do a calculation based on the real world data, and just add 8% on top -- that will average out over an 8hr flight or so. If your flight is less than 8hrs, you will probably end up with less fuel than planned. If your flight is longer than 8hrs, you will likely end up with more than planned.

 

I'm not saying it's impossible, and if you recorded lots of data at different weights, altitudes, TAS etc you could probably create a very accurate profile, and if I were the OP writing a spreadsheet from scratch that's probably what I would do. But refining a profile from the real-world data is difficult because the relationship between the real-world figures and the PMDG 747 actual figures over the course of a flight is distinctly non-linear.

[edgarvsfreitas 0]

Sorry, I didn't make myself very clear.

 

What I meant was that about +8% (in Simbrief) and about +9% (in PFPX) worked well for me in terms of giving me a sensible overall fuel load.

 

However, the problem with creating a truly accurate profile for the PMDG 747 (if you're basing it off real-world data) is that it is not as simple as saying that "it burns 8% more than the book says", because the amount it burns more than it should varies over the course of a flight. So 8% (or 9%) is an average figure which will get you in the ballpark for a typical stage length of about 8hrs.

 

However, from what I recall the PMDG 747 burns far too much fuel (maybe book +10% or +12%) in the early stages of the flight. As the flight progresses, the fuel burn (as expected) decreases, but not at the same rate as the fuel burn decreases in the real thing. So after 4 hours, perhaps, it might be burning book fuel +8%. After 8 hours the fuel burn will have decreased further, to perhaps book + 4%, and so on.

 

So if you do a calculation based on the real world data, and just add 8% on top -- that will average out over an 8hr flight or so. If your flight is less than 8hrs, you will probably end up with less fuel than planned. If your flight is longer than 8hrs, you will likely end up with more than planned.

 

I'm not saying it's impossible, and if you recorded lots of data at different weights, altitudes, TAS etc you could probably create a very accurate profile, and if I were the OP writing a spreadsheet from scratch that's probably what I would do. But refining a profile from the real-world data is difficult because the relationship between the real-world figures and the PMDG 747 actual figures over the course of a flight is distinctly non-linear.

Dear Simon,

 

By far that was the most complete and clarifying answer I've got since I decided to post this question, not only here but in other websites as well.

I've made the spreadhseet by scratch, studying the manuals and doing it step by step. I don't intend to make it as the real 747, as I don't have enough knowledge to do it, but, I was intending to make it as the Felice Valente's fuel calculator. Why? Because for me, it's the most accurate out there, and gives you the exact amounts, with very low error. I don't know if you know this fuel calculator but I strongly recommend it. I've noticed that the calculations made exactly as the book gives you less fuel than you need to accomplish a flight, because I've made a few flights using those amounts and I finished with around 5000lbs in the tanks. 

I made a few changes in the fuel calculator and now I get sometimes +20000lbs of fuel, which can be good, because like you say, the pmdg 747 burns more fuel than what's expected!

I dunno if you are interested in taking a look to the spreadsheet I've created and maybe in giving me an opinion, but if you are, please send me a message and I'll send it to you.

 

Once again, thank you so much for your answer, it helped a lot!

 

 

Kind regards,

 

 

Edgar Freitas

[downscc 1,327]

 

 

However, from what I recall the PMDG 747 burns far too much fuel (maybe book +10% or +12%) in the early stages of the flight. As the flight progresses, the fuel burn (as expected) decreases, but not at the same rate as the fuel burn decreases in the real thing. So after 4 hours, perhaps, it might be burning book fuel +8%. After 8 hours the fuel burn will have decreased further, to perhaps book + 4%, and so on.

 

I do not remember the specifics from my sessions I ran over ten years ago, but this is at least analogous to my experience which is why I had to create a spreadsheet with segmented functions and run a linear regression on actual measurements.  Haha, this was my second PMDG aircraft and I really dug deep into it... it was fun but not sure I'd do it again.

[edgarvsfreitas 0]

I do not remember the specifics from my sessions I ran over ten years ago, but this is at least analogous to my experience which is why I had to create a spreadsheet with segmented functions and run a linear regression on actual measurements.  Haha, this was my second PMDG aircraft and I really dug deep into it... it was fun but not sure I'd do it again.

I've seen your spreadsheet Dan. Looks quite good and I would like to insert a linear regression on the flight plan fuel but dunno how to do it. That could probably solve my problem by reducing the fuel load when the total fuel is too much and increasing the fuel when is too low.

[downscc 1,327]

Linear regression is a method used to reduce measurements to a best fitting line on a graph... I may have even used a higher order regression in this case.  This has nothing to do with the use of the final product and everything to do with how it was created.

 

Linear regression was the brainchild of the worlds greatest mathematician Gauss.

[edgarvsfreitas 0]

Linear regression is a method used to reduce measurements to a best fitting line on a graph... I may have even used a higher order regression in this case.  This has nothing to do with the use of the final product and everything to do with how it was created.

 

Linear regression was the brainchild of the worlds greatest mathematician Gauss.

I probably wasn't clear. The only number that doesn't match in the calculations is the flight plan fuel (no contingency fuel, alternate or MLF added here), with a linear regression I could refine that number when it's to high or too low. I know why and when it is high/low, the problem is making that "refinement" on a spreadsheet. But I'll look about how to make a linear regression on excel. Thanks Dan.

[Johan_Dees 84]

An old and late friend and real 744 captain Mel Ott, learned me that the -400 used 22000 lbs and hour average.

So 10 hrs would make 220,000 lbs, and would be a catch all. (25000 for the -200).

He checked the predictions with this and they should be close.

Works for me all the time.

 

http://www.avsim.com/pages/contrails/contrails_ott3.shtml

 

Mel was an early PMDG member.

[phil747fan 14]

An old and late friend and real 744 captain Mel Ott, learned me that the -400 used 22000 lbs and hour average.

So 10 hrs would make 220,000 lbs, and would be a catch all. (25000 for the -200).

He checked the predictions with this and they should be close.

Works for me all the time.

 

http://www.avsim.com/pages/contrails/contrails_ott3.shtml

 

Mel was an early PMDG member.

Johann,

 

those days when we were young are long gone...

 

yes Mel was very good and easy to answer all the dummy questions we had in those days.

 

do you remember when he wrote the story about doing a roll over in the sim?

 

i may have to compile one day the datas i have for doing fuel planning on 747s ...

 

Phil.

[Johan_Dees 84]

Hi Plillipe, another great friend of those Vector's days!

 

Yes, those days are long gone, but not in my mind.

 

I remember the roll story well. I miss Mel.

 

Somewhere I have the Vectors Q&A stored, if I find them I will upload them. :-)