Refueling behavior question

[donaldseadog]

...

So for a group with a single unit receiving fuel you can what I call cheat and split the tanker as soon as you have got the jump in bingo fuel, the increase in range ring or the change of "with refuel" to "no refuel", but to do so is cheating the simulation. If you try this with a multi receiving unit group you'll leave only the first unit with new fuel, the remaining units as they were and an 'empty' tanker cruising home of no further use.

...

 

What sort of implications would this condition produce? For instance, would the range circle now be bogus - or, if not, then which unit would it apply to: the one that got fuel or the others? Also, how would it affect the actual range of the overall group? Would some of the members of the group eventually crash, while leaving the re-fueled member still flying (given they don't reach base before that, of course)?

I think we might be confusing terminology. I use the term unit as in the harpoon sense, a group is made up of at least one unit, each unit (in the case of aircraft) is made up of at least one individual of same type/loadout - you see units in the unit window, you get information on them in the unit report table, etc. I'm not talking individual planes, for this sort of thing harpoon seems to only work with the unit, not individual aircraft.

 

In refuelling each unit is refuelled as a single entiity and it occurs instantly, if there are two units there is a time gap (simulating the real time to refuel aircraft) between the refuelling commencement times of the respective units, in this gap you can split off the tanker and the later units don't get fuel, (this is what I'm saying) as would be the case in reality.

Don

[donaldseadog]

Here are some results from a refueling test.

 

This is not by any means an exhaustive test, either by considering all variables, or by considering any variable completely. The air groups were formed up and sent to a distant patrol location. Fueling happened automatically in all cases.

 

Note that air groups contain at least one unit, each of which has at least one plane. All planes in a unit are identical, except that they can have different numbers of weapons remaining.

 

1) 1 x F/A-18C plus 1 x F/A-18C tanker. At some point, the message log stated that the refueling process was beginning. 30s later, the message log stated that a particular unit was being refueled. 3 min later, the tanker asked to be sent home.

 

2) 5 x F/A-18C plus 2 x F/A-18C tanker. The group consisted of 3 units: one with 2 tankers, one with one regular Hornet, and one with 4 regular Hornets. At some point, the message log stated that the refueling process was beginning. 30s later, the message log stated that the first unit (1 plane) was being refueled. 1.5 min later, the message log stated that the second unit was being refueled. 6 min later, the tanker asked to be sent home.

 

As soon a unit starts refueling, its fuel status jumps, i.e. the fuel is transferred instantaneously. There is a significant delay before the next unit gets its fuel.

 

These numbers are completely consistent with all previous results in this thread, except Brad's. But Brad's results are also consistent if he did not report the 3-min delay after fuel transfer and tanker splitting. This means that type of tanker and fuel transfer rate are not modelled.

 

Summary:

 

Time to refuel a plane unit = 3 min x number of planes / number of tankers. Because the fuel is transferred instantaneously, you can cheat by splitting the tanker off before the final delay period is complete. There will be cases where this does not make sense, because in RL, only one tanker can refuel a plane at a time.

 

Incidentally, when launching the 7-plane group, because the patrol point was beyond the range of the basic Hornet, I had to add the planes in the following order: tanker, Hornet, Hornet, tanker, Hornet, Hornet, Hornet, after which another tanker would have been required. The conclusions are obvious.

 

It also appears to be the case that the "bingo"status in the group window is the same as the "bingo" status for the first unit in the unit window.

Nice summary,

Brad has pointed out that the database doesn't distinguish tanker characteristics (although there is the actual range value that differs for each tanker class) so there isn't much room to fine tune things like delivery period but for harpooon purposes I'd think the current is quite good enough.

There do appear some inconsistencies though, I say appear as I haven't bothered to look at them empirically:

When you put a group together for a flight plan that requires tankers then the total 'range' value of the tankers seems to have to be equal to or greater than the total range deficiency of the non tanker units, implying that the range vale for each tanker is taken as a kind of value for the fuel available. However I remember being told, or having read somewhere that the available fuel is twice the range of the tanker and that does seem to correspond to what I see 'in the air'.

Also the time at which refuel automatically takes place, I'm sure it used to be at about 25% remaining fuel of the 'thirstiest' unit but now it seems to occur seudo random (ie I don't know the criteria). My best guess is that it is the earlier of: 25% remaining fuel; ability to take all the tanker fuel on offer( that would be very good criteria). There were changes to code for IFR about two years ago to solve tankers crashing in long ferry trips and poor distribution to multi unit groups.

 

Anyway it might be good to work out rules when auto IFR commences and how much fuel a tanker provides.

Using the tanker logging it looks like it is the tanker range (also at group assembly for launch) but I'm certain I've seen groups get more than that. Logging doesn't seem to help regarding calculating when to commence auto IFR.

[CV32]

Brad has pointed out that the database doesn't distinguish tanker characteristics (although there is the actual range value that differs for each tanker class) ...

 

Yeah, but its only the range contained in the loadout that matters.