[

For certain long-range missions, especially involving multiple refueling, it will be necessary to ensure that the tankers are in individual groups, because you will probably want to run the tanks dry before manually requesting more fuel, and the automatic procedure doesn't do that.

 

Yep, they way I use is to set up several patrols to the target area with enough tankers to get to the next fuel stop and then some (one never knows when the use of 'burners will be advisable). sometimes it's possible to have the first patrol return to base, refuel and then take off again to support the return flight.

In what appears to be becoming a never-ending quest to understand the refueling process, I ran a few more tests and got some more anomalies.

 

1) Delay times

 

(No anomaly in this section, just more details.)

 

It appears to be the case that the critical factor deciding when IFR is to occur is when a plane is "low on fuel". The "low on fuel" check is done at 30 s intervals, on the minute and half-minute. If the tanker starts not grouped with the plane to be refueled, then the time between grouping and the next low fuel check must be added to the delay time. After that, there is a 60 s delay for the refueling process to start, and then there is a further 30 s delay for the refueling process to be finished.

 

For example:

 

30 min 12 s - group planes

30 min 30 s - check fuel (invisible, you don't actually get notified of this, it's still a guess that this happens)

31 min 32 s - start refueling (additional 2 s delay for computer processing time)

32 min 2 s - refueling is finished

 

If the planes were already grouped, the first 18 s delay would not exist.

 

The above applies for automatic refueling. If you force refueling with ALT-r, then there is a 30 s delay to start the refueling process, and then a 30 s delay for it to be finished.

 

For example:

 

30 min 12 s - ALT-r (you get a confirmation message, sometimes preceded by a confirmation request)

30 min 42 s - start refueling

31 min 12 s - refueling is finished

 

2) What is "low fuel"?

 

As of post 15, this thread, I thought that "low fuel" meant that the plane in question was at least blinking. At that point, and even at 0% bingo, the plane still has enough fuel (wood or coal) to make it back to base on its own steam. But I now have at least one concrete example where the automatic refueling process would not start until the plane was no longer able to make it back to base with what remained in the tanks.

 

Further investigations are indicated.

Some measurements for "low fuel" status

 

At what point does an airplane call in a tanker?

 

In order to try to get a handle on this question, I ran the following experiment:

 

1) Loiter a plane somewhere at some altitude.

2) Loiter a tanker nearby at a different altitude.

3) Group the two.

4) Wait, possibly increasing the clock rate.

 

The different altitudes will hopefully provoke a prompt before refueling happens, so that the measurement can happen just before the event. Although an altitude difference is corrected if the group is moving, this does not happen if the group is loitering.

 

What we learn:

 

1) At 10% above bingo fuel, you get a warning that the plane is approaching bingo fuel. This warning is suppressed if the plane is grouped with a tanker.

2) At 5% above bingo, the plane group starts to blink.

3) At 0% bingo, the plane will ask to be sent back to base, but not if the plane is grouped with a tanker.

 

For the specific situations:

 

Scenario is USNI 4 BLUE. The tankers are Vikings.

 

1) F-14A 185 nm from base refuels at 275 nm fuel remaining.

2) A-6E 235 nm from base refuels at 235 nm fuel remaining.

3) F/A-18A 310 nm from base refuels at 230 nm fuel remaining.

4) S-3B 390 nm from base refuels at 425 nm fuel remaining.

 

Seems to be all over the map.

 

If anyone can explain these results, please add your comments.

This is turning into a good examination. I'll give a little divination from the code as a bonus to the discussion.

 

One of the automated checks to create a refueling event happens every 60 game seconds for every group or immediately for the current unit or group if a staff report is requested for the selected unit or group.

 

Within that check a refueling event is created if the group is below 50% bingo fuel and there is a tanker in the group (and there isn't already a refuel event for the group) and the group is fully airborne (all .

 

Now just because a refuel event is placed it doesn't mean a refuel will happen in the next static ### of seconds, just that a finer tuned check will happen and decide whether to start refueling and if not, requeue for later action.

 

That's it for now, keep at it folks.

This is turning into a good examination. I'll give a little divination from the code as a bonus to the discussion.

 

One of the automated checks to create a refueling event happens every 60 game seconds for every group or immediately for the current unit or group if a staff report is requested for the selected unit or group.

 

Within that check a refueling event is created if the group is below 50% bingo fuel and there is a tanker in the group (and there isn't already a refuel event for the group) and the group is fully airborne (all .

 

Now just because a refuel event is placed it doesn't mean a refuel will happen in the next static ### of seconds, just that a finer tuned check will happen and decide whether to start refueling and if not, requeue for later action.

 

That's it for now, keep at it folks.

 

Hi guys, having spent quite some time, like all of you it seems, playing with refuelling scenarios etc I think it is cheating to be able to look at the code, no wonder the group is called "staff pukes" (only joking).

Two traps not mentioned here that I sometimes forget is that if your tanker has less range than the recipient, you some times run the tanker dry on the way home (oops), the other, when ferrying over a long distance if the group includes a tanker at take off, refuelling may occur about 30 secs after group formation (takeoff completed) and thereby having no effect (otherthan having to bring every one back and refuel on land).

You wouldn't want it all to be too easy though, would you?

Don.

This is turning into a good examination. I'll give a little divination from the code as a bonus to the discussion.

 

One of the automated checks to create a refueling event happens every 60 game seconds for every group or immediately for the current unit or group if a staff report is requested for the selected unit or group.

 

Within that check a refueling event is created if the group is below 50% bingo fuel and there is a tanker in the group (and there isn't already a refuel event for the group) and the group is fully airborne (all .

 

Now just because a refuel event is placed it doesn't mean a refuel will happen in the next static ### of seconds, just that a finer tuned check will happen and decide whether to start refueling and if not, requeue for later action.

 

That's it for now, keep at it folks.

 

Hi guys, having spent quite some time, like all of you it seems, playing with refuelling scenarios etc I think it is cheating to be able to look at the code, no wonder the group is called "staff pukes" (only joking).

Two traps not mentioned here that I sometimes forget is that if your tanker has less range than the recipient, you some times run the tanker dry on the way home (oops), the other, when ferrying over a long distance if the group includes a tanker at take off, refuelling may occur about 30 secs after group formation (takeoff completed) and thereby having no effect (otherthan having to bring every one back and refuel on land).

You wouldn't want it all to be too easy though, would you?

Don.

 

Let me add some comments from a rather different perspective.

 

Instead of trying to figure out how the automatic refueling actually works (and the more I learn, the more confused I get), I will say the way I think it SHOULD work.

 

On any given aircraft mission, the pilot (and probably a few other people) will know how much fuel is needed for normal completion of that mission. Let us consider only those missions where at least one refueling event will be needed, but no more than one. Let us not bother with how the tanker will get there. The tanker will be teleported in as needed.

 

The refueling event can occur any time between the time the plane's tanks are sufficiently empty to hold the required additional fuel and the time the plane is totally out of its original fuel (well, with a small reserve). I can see how it might be a good idea to trigger the refueling event as soon as possible, in order to release the tanker as soon as possible. So if you include a tanker with a plane group, the program can calculate how much fuel will be transferred and start the transfer as soon as there is sufficient space in the recipient's tanks. If you want to transfer the fuel earlier (because you do not need the full amount), you could use ALT-r. If you want the fuel later, you could keep the tanker in a separate group.

 

The main problem with this approach is it requires a multi-unit calculation, which is more complicated than the single-unit calculation I have been using so far. That is to say, it is not just the fuel state of the recipient that initiates the transfer process, but the state of that plane AND the tanker (and maybe other planes as well).

 

What I don't understand is why 50% bingo would be used, because that occurs when the plane has used up maybe 20% of its initial fuel.

 

I also don't understand why refueling can occur immediately after launch. Surely that is just a bug (i.e. code written to do something the code writer didn't want) which has been repaired by now?

Hi guys, having spent quite some time, like all of you it seems, playing with refuelling scenarios etc I think it is cheating to be able to look at the code, no wonder the group is called "staff pukes" (only joking).

Two traps not mentioned here that I sometimes forget is that if your tanker has less range than the recipient, you some times run the tanker dry on the way home (oops), the other, when ferrying over a long distance if the group includes a tanker at take off, refuelling may occur about 30 secs after group formation (takeoff completed) and thereby having no effect (otherthan having to bring every one back and refuel on land).

You wouldn't want it all to be too easy though, would you?

Don.

 

Hehe, we do like to poke fun at ourselves on occasion. Looking at the code often is cheating, refueling however is so convoluted that looking at the code doesn't get a person all that far, the guess and check method of testing produces far more useful results in this case (though if you guys keep up the good discussion I may keep ducking my head into the code and trying to interpret what it is doing).

 

The refueling 30 seconds after takeoff is one thing I tried to squash between 2008.044 and 2009.036 and it happens much more rarely now than it used to occur. Someone could make a repeatable (small, alwyas small) test scen showing such a situation and make a new Issue Report .

Hi guys, having spent quite some time, like all of you it seems, playing with refuelling scenarios etc I think it is cheating to be able to look at the code,

 

Looking at the code often is cheating,

 

Why? If it produces the answer quickly, it should be the recommended approach.

 

refueling however is so convoluted that looking at the code doesn't get a person all that far, the guess and check method of testing produces far more useful results in this case

 

That's useful to know, because it means we should NOT be looking for simple explanations, i.e our efforts to come up with simple explanations are bound to give wrong explanations.

I think this is one of those situations that will defy solution, sometimes it will be necessary to fly a mission to nearly empty tanks (and I had the case when refueling after joining a group couldn't be arranged fast enough ) to get to a target so that will require multiple refuelings with tankers cycling in/out of home base to get the strikers to base.

Not sure what the problem is for you folks. I get 100% mission capable rate out of my HCE tankers. Go ask the USAF (and for that matter, any air force) if they're doing as well.

Not sure what the problem is for you folks. I get 100% mission capable rate out of my HCE tankers. Go ask the USAF (and for that matter, any air force) if they're doing as well.

 

Yeah, well, THAT can be tackled by the designer by cutting back available airframes by the average unavability rate and then some of the rationale for the F/A-18E will be apparent...

Yeah, well, THAT can be tackled by the designer by cutting back available airframes by the average unavability rate and then some of the rationale for the F/A-18E will be apparent...

 

zing!

On the idea of just plain starting over with refueling (as mentioned earlier in this topic), look to http://harpgamer.com/harpforum/index.php?showtopic=2233 for some potential ideas on capabilities that require little disruption otherwise to the game.

I have mentioned some errors in the refueling procedure earlier in this thread. In this post, I provide some more careful measurements to demonstrate this.

 

The scenario is USNI 3 (RED). The Russian airbase is Olenya. It has one runway.

 

The tanker being used is the Tu-16N Badger A. It has a normal range of 3100 nm. It cruises at 460 kn at medium.

 

The test plane is the Tu-16K Badger G. With an ASuW loadout, it has a normal range of 2480 nm. It cruises at 460 kn at medium.

 

The general procedure I used was to launch one bomber at the start of the scenario, then launch one tanker along the same path, then group them. This separates the planes in the unit window for easy identification.

 

At 460 kn, the planes consume just under 8 nm of fuel per minute. Of course, they are also travelling, but that doesn't matter here.

 

At specific times, I forced refueling (ALT-r). The planes were never actually low on fuel, and automatic refueling did not occur.

 

After each measurement was done, I restarted the game for the next measurement.

 

At 5 min, the bomber had 2444 nm of fuel left, and I forced refueling. After 1 min, its fuel went to 3028 nm.

At 10 min, the bomber had 2404 nm of fuel left, and I forced refueling. After 1 min, its fuel went to 3037 nm.

At 15 min, the bomber had 2366 nm of fuel left, and I forced refueling. After 1 min, its fuel went to 3037 nm.

At 30 min, the bomber had 2251 nm of fuel left, and I forced refueling. After 1 min, its fuel went to 3037 nm.

 

We can see that the bomber is using about 38 nm of fuel per 5 min, which is correct. Fluctuations can be accounted for by reaction time. The 5 min final fuel status is anomalous, but that may be due to a bad measurement. The other numbers are consistent.

 

The main problem with these results is that the bomber is getting more fuel than it can actually carry.

 

For the next measurement, I used 10 bombers in a group, but otherwise used the same procedure as above. Since there were only 8 of these bombers immediately available with the ASuW loadout, I used the same bomber, but these ones had the ASuW-LR loadout. The range is still 2480 nm, and for purposes of this test, I expect that there is no difference. Of course, it takes about 5 min to launch the 10-bomber group.

 

At 15 min, the bombers had 2375 nm of fuel left, and I forced refueling. After 1 min, their fuel went to 2927 nm.

At 30 min, the bombers had 2260 nm of fuel left, and I forced refueling. After 1 min, their fuel went to 2812 nm.

 

In the first 15 min, the bombers used 105 nm of fuel. This is slightly less than the single bomber, due to the loitering during launch. In the second 15 min, the bombers used 115 nm of fuel, which is correct.

 

We should now have a fairly good handle on the fuel transfer process.

 

Since I recorded the bomber fuel status before and after the refueling process, which takes 60 s for forced refueling, we need to correct for the fuel used in that time, which is 8 nm.

 

The 10-bomber group got 560 nm of fuel (x 10).

 

Because this value is the same for both tries, and the bomber final fuel status is not constant, we know that this is the maximum amount of fuel one tanker can give.

 

For the single-bomber tries, the final fuel status is constant, and the amount transferred is not constant, so we know that 3037 nm is the maximum fuel the bomber can hold (with this loadout).

 

So then I decided to launch all 24 Badgers at once (16 ASuW-LR, 8 ASuW). For the first 4 minutes, they were launched in pairs. It took 4 minutes to launch the remaining 8.

 

At 10 min, the bombers had 2419 nm of fuel left, and I forced refueling. They had used only 61 nm of fuel, considerably less than the normal 77 nm; this is due to the extended launch loiter time. After refueling, the 8 ASuW bombers had 2651 nm of fuel, but the 16 other bombers got nothing! After correcting for the transfer time, we can see that 240 nm (x 8) was received. However, if all the fuel is going to only 8 planes, then they should have received 5600 nm / 8 = 700 nm each. 5600 / 24 = 233, so the tanker gave away all its fuel, but 2/3 of it was simply dumped into deep space!

 

Finally, I used the available group of 24 Su-24M Fencer Ds. These planes cruise at 561 kn at medium. I launched them all in one group, which took 12 min. I put them on afterburner for a while to use up fuel. Then I loitered them at high. Then I brought up one tanker, loitered it at low, and added it to the group. Then I forced refueling, using anti-gravity feed and a VERY long hose.

 

The instant before the refueling was completed, the Fencers had 467 nm of fuel. The instant after the refueling was completed, the Fencers had 708 nm of fuel. Total fuel transferred was 241 nm x 24 = 5784. I can't account for the extra 184 nm of fuel.

 

Interesting detail : 184 / 24 = 23 / 3 = 460 / 60. On the left, this is equal to the unexplained additional fuel each Fencer got. On the right, it is equal to the fuel a Badger (bomber or tanker) uses in one minute at cruise.

 

Problems with the details of the process:

 

1) The Tu-16N Badger A tanker appears to hold 5600 nm of fuel for transferring. This is much more than its normal range of 3100 nm and may not be correct.

 

2) The Tu-16K Badger G can be refueled up to 3037 nm, which is about 22% higher than its normal range with this loadout. But its normal range probably expresses how far it can fly with tanks full, so the 3037 nm is probably incorrect.

 

3) When refueling a large group of planes from one tanker, if the planes are different by as little as having a different loadout (and I expect that it may be possible to reduce this difference to something even smaller), then the full number of planes is used to calculate how much fuel each plane gets, although only some of them actually get the fuel.

 

4) There may be a problem with determining how much fuel a tanker can give, because it seems to depend on which plane is getting the fuel.

 

5) If you are launching planes from a base with a limited number of runways, then the 30 s per launch per runway restriction is adhered to, as long as you are launching only one type of plane. But if the planes are different by as little as having a different loadout, then each different type uses the runways independently, so the effective number of runways is the actual number of runways multiplies by the number of different plane types. This is not a refueling bug, and it is not a new issue, but it is a mistake in the program, and players will need to take some care to avoid exploiting this (unless they like cheating).

More refueling data.

 

Harpoon version 2007.000.

 

GIUK 1 (NATO)

 

refueling EA-6B Prowler from A-6E tanker

 

1) from 25 nm of fuel to 1763

2) from 36 nm of fuel to 1763

3) from 199 nm of fuel to 1763

4) from 0 nm of fuel to "tried to refuel a unit but it did not exist". Oh well, can't win them all.

 

As in the previous post, we see that if a lot of fuel is available (the entire tanker load for one plane), a variable amount of fuel is transferred, up to the total capacity of the refueled plane.

 

The problem is that the Prowler can only carry fuel for 1440 nm. For reference, the A-6E tanker range is normally 1800 nm.