FSX Fuel Truck Behaviour Around Parked AI Aircraft

[dickb]

I've been having some fun adding crew buses to my WW2 airfields using the fuel truck system. I've replaced the default sm2 fuel truck model with one of Ian P's RAF trucks and set up the airfield with enough Vehicle parking spots to service my AI aircraft. And it all works pretty well - the buses drop off crew before startup and (I think - I haven't tested yet) collect them after landing. They use the main taxiways, but so far I haven't seen any major conflict with AI traffic.

The only problem is what the buses do when they arrive at the AI parking spot. What I'd like them to do is pull up in front of the a/c or drive round to the back. What they actually do is drive diagonally right to left across the parking spot, straight through the AI model and out the other side, where they lurk somewhere off the starboard quarter. After a while they back up, and then drive back through the starboard tail plane and undercarriage to rejoin the taxiway.

Since most of the parking spots (dispersals) are remote, and most of my AI aircraft are heavy bombers, this odd behaviour is generally not obvious, but I'd like to improve it if I can.

Any ideas ? Parking spots are all set up as PARKING, Radius 75ft.

Dick B

 

[raymk]

Have you tried adding some specific "Vehicle" type links so that, firstly, your vehicle don't use the main aircraft taxiways and, secondly, so that, in my experience, the vehicle will get to the nearest taxi node (which is usually behind the parked aircraft) and then move out to diagonally behind the aircraft on the starboard side?

I guess the stopping point distance is controlled by the parking radius but I have actually tested this myself.

 

[dickb]

I don't really want to add a completely separate vehicle network if I can avoid it. Apart from the fact that it would be a lot of work, it's actually (IMO) more realistic for these vehicles to use the taxiways, and the way I have it set up they are only on the main taxiway (perimeter track) for a very short distance, so the chance of conflict with aircraft is relatively small. That said, I will experiment to see if I can direct them around the side/back of the parking spots with a short vehicle route off the taxiway. I'm guessing the vehicles will ignore that, but you never know !

I should probably explain that the parking spots I am working with at the moment are of the circular 'banjo' dispersal pan variety, where the entrance/exit taxiway enters the pan at the front, so the nearest taxi node is at the front of the a/c.

Size of parking radius doesn't seem to be a factor. I doubled the size of a parking spot and didn't see any appreciable difference.

 

[raymk]

Just completed a fairly quick test on an airport that I am currently working on.

Put some FA-18's on spots of radius 8 metres and 15 metres with the taxiway pointing out the front of the spots. When the fuel truck approached from the front on the latter spots, it avoided the aircraft and pulled off to the starboard rear. It then reversed back to the aircraft, just hitting it, before driving away. On the smaller spots, it passed thro the aircraft before again going to starboard rear (but not so far away). It then reversed thru the aircraft before driving off.

So I suspect that the size of spot does have a bearing on the vehicle's behaviour but this wasn't a totally scientific test.

 

[dickb]

Interesting - I'll have to take another look at that. Maybe my brief test was flawed. Or maybe it has something to do with the geometry of the a/c versus the parking spot.

Unfortunately my development PC has just gone belly up (suspect motherboard), so I may be a while ! Thanks for the input though. I'll report back in due course.

 

[gfxpilot]

I'm not 100% on this but it looks like "buses" are stopping using the same method as the default fuel trucks. I noted in a rebuild of lowi I'm polishing up. After setting the vehicle tracks and setting parking I noted the truck approach and stop starboard side at about 5 o clock to the flight deck. If I changed and parked in as smaller spot with the same 737 the truck cut up the tail .

 

[dickb]

I'm not 100% on this but it looks like "buses" are stopping using the same method as the default fuel trucks. I noted in a rebuild of lowi I'm polishing up. After setting the vehicle tracks and setting parking I noted the truck approach and stop starboard side at about 5 o clock to the flight deck. If I changed and parked in as smaller spot with the same 737 the truck cut up the tail .

As far as FSX is concerned, the buses are fuel trucks - just a different model. What you're seeing tends to confirm raymk's observation about the size of the parking spots. Also, if I read you correctly, your trucks arrive at the parking spot via vehicle tracks rather than taxiways, and yet the result is much the same, so that's one option I probably don't need to explore. Out of interest, in the cases you mention, where does the vehicle track join the parking spot relative to the aircraft ?

 

[gfxpilot]

The track does not actually "join" the aircraft's parking spot. The vehicle has its own parking node . if you look at ADE. For vehicle to follow the vehicle track and not use the taxiways you need to make sure all the blue nodes are connected (the blue dots) if there is a break at all between the blue nodes the vehicle will opt for a taxiway and ATC gives the vehicle preference over aircraft on that taxiway (strange but true when you see an A1 747 stop for sewage wagon). The fuel truck (bus) will follow the following rules. It requires the aircraft to be parked correctly in the right size aircraft parking node (green dot) a 737 parked in a " heavy" parking spot will not get refueled. A 737 parking a in small space will see the fuel truck hit the 737,s tail section. From my experience each aircraft parking node has the refuel truck park at a preset distance to the rear of the aircraft on it starboard side (adjacent to the rear horizontal stabiliser right side).

When you call the bus " fuel truck" you'll be aware that its arrival will determined by the plan being in the right size parking zone,aligned correctly etc. I haven't attempted to bring a vehicle node ( blue dot) closer to a parked aircraft as yet but I'll give it shot tomorrow. One thing to remember. Each time the bus arrives you'll fuel to 100%. If you fly using any online community which tracks fuel hankering as as cost feature you'll have to readjust your fuel carry or you'll be taking off heavy everytime

 

[dickb]

I'm confused. Are you talking about user aircraft or AI aircraft ? Or both ? At the moment I'm only concerned with AI aircraft.

I know the truck/bus has it's own vehicle parking spot from which it starts. Are you saying it also has its own parking spot/node at each aircraft parking spot ?

 

[dickb]

Whoopee - got my development PC working again, so I can test some of this stuff myself !

I did just take a look at the ADE manual, and the example there does show the vehicle path joining the aircraft parking spots (at the rear) via an apron link. See attached pic. The same link that the a/c uses. My AI a/c enter and depart from the front of the parking spot. Can a parking spot have two links I wonder ? Something to try !

 

[dickb]

Well - after several hours of experimentation it looks like I'm just going to have to add a dedicated vehicle network. No matter how large/small I make the parking spots, buses/trucks using the main taxiway drive straight across the middle and through the aircraft. Fortunately I don't have to cover the whole airfield, just the dispersals for A and B flights, so not too bad. I plan to run the vehicle tracks (zero width) along the edge of the main taxiways. Hopefully that way they won't get mown down by any taxiing a/c.

Thanks all for your input !

 

[gfxpilot]

I've been having a rumble using the fuel truck. . first apologies for not being clear about the vehicle link connecting to the aircraft parking (green circle) , I meant to make it clear that the vehicle link does not connect to the green parking circle directly but to the taxi node. It appears the truck needs the vehicle path set one side of the taxiway. You'll have the vehicle path then along side the taxiways. When I connect from the vehicle to the taxiway I made the link between the parking spot and the parking spots junction at the taxiway. I changed the parking spot to 10metre for the test. I set the T1 in the parking spot offset at 15 metres. That lines a 737 up reasonably well. I then made the vehicle parking spot nearby and connected to the vehicles pathway. I tested that in Ade and it worked . the picture you attached does not shoe the taxiway and vehicle path seperatley

 

[dickb]

Could you possibly attach a screenshot of your test in ADE ? I'm not sure I follow your description accurately.

 

[gfxpilot]

yep Ill get it sorted asap for you

 

[gfxpilot]

Here is what I have thus far - Lowi Ramp three allocated in image 1 = 10 mtre parking spot. Along side to the picture right is the parking spot for the Fuel (our your bus) you can just make out the purple circle beneath the yellow information box.

Image 2 - How it looks live before (Shift -F) fuel is called

Image 3 A- Having called the truck does a 3 point turn and then travels up to the taxiway turns and then starts down the Aircrafts taxiway behind the rear of tail and maneuvers under the tail stabaliser and parks between both starboard wings. as image (remember thats with a 10 metre paring sport with "T1" set to 15 )
Image 3

I then increased the parking sport from 10 to 25 metres (Image 4) - recompiled and launched FSX. I followed the same process as images 1 , 2 3A. The fuel truck still followed that same path as in 3a but this time it stopped way to the right of the tail stabiliser (image 5)

From my trials here to day and some added reading it appears that its always better to have a vehicle path to the side of a taxiway, that minimises the sewage truck stopping a 747 in its tracks because they share the same space.

Where a vehicle node (Purple track blue dot) is missing in the trucks expected path FSX has been promed to use the nearest taxiway point it can find to get to the calling Aircraft - this does work occasionally but is not guaranteed.

You will see Ive connected the vehicle tracks (Purple track blue dot) with the taxiway link coming to the parking node not directly to the node itself. This seems to work better. Ive also looked at where the best and most efficient parking spot was for the vehicle - In my images the one truck happily serves both ramp 3 and 4 in an identical manner

As a trial you will see directly in front of the tower I have a parking spot for a vehicle (Image 1) to the right of that again I have the Cargo Ramp with eye candy and parking flagged as large (45mtres) this is where Lowi has it cargo hangers presently. I removed the vehicle parking note (purple dot) . I then recompiled and spawned at that ramp with the same ups 737. When I called the fuel truck the closest one was between ramps 3 and 4. It took a while but the truck finally came and parked as expect a considerable way off the tail stabilizer to the starboard side.

Conclusion - make sure that at your AC parking (green node) is set to as small as it will allow i.e. 10m min for a 737 if its smaller then I found that the truck does not react at all.

Check that your vehicle path is added AFTER you place your taxiways and AC parking so that you make it as efficient as possible -
Be ware that you are using the fuel call to bring a bus to its better to have the bus parking as close as you can to the Aircrafts parking area that a bus will be allowed at.
Remember that when the bus arrives it will trigger a fuel fill up as well.

The issue of the fuel truck hitting mid to rear fuselage may have a lot to do with the "T" settings in the aircraft parking node because if T1, 2 3 and 4 are all at zero the aircraft will prak/spawn more or less on the green tee line in "image 1" of the green parking node which is quite a way further back than you want and the fuel truck uses the rear edge of the green parking node as its target to cross and park. Offsetting pulls the front wheel of the aircraft forward. - Might be wise to trial the "T" offsettings at different distances but I found 15 in T1 is ok.

Im no expert (at all) on this but the points above all were worked through with copious amounts of galley visits and the odd Chocolate biscuit for good measure.

Let me know how you get on -its an interesting point

 

[dickb]

Hi GVX pilot - thanks for all this - great stuff !!!

However, I see one major difference between your airport layout and my WW2 airfields. I probably should have posted a pic before, but here is what one of my typical parking spots (dispersals) looks like:

You will see that unlike a civilian airport, the aircraft enter/exit the parking spot at the front, not the rear. I also need the bus/truck to enter/exit the parking spot via the same route. It's not particularly realistic (though not unknown) to have them going offroad, particularly in inclement weather.
This is not a nice big concrete apron like you airline (and modern military) folks have to play with !

I have already established that using the taxiway doesn't work - as mentioned above, the bus/truck just drives through the parked a/c to get to its favourite spot to the rear quarter. Somehow I have to get it to go round. Increasing the size of the parking spot doesn't seem to make any difference either (for me that is).

I eventually decided to put in a separate vehicle route overlaying the existing taxiways and skirting the parking spot. This is fine provided it doesn't actually intersect with the taxiways at any point. It can cross them, but not intersect, otherwise the bus/truck can switch to the taxiway if the route is shorter. Width is zero so it's invisible. (EDIT: This is more important than I realised - if the vehicle track has any width, even though it doesn't look like it intersects with the taxiway, the bus/truck thinks it does, and can move from one to the other.) Notice the separate vehicle path needs its own link to the parking spot, otherwise the bus/truck won't attend:

This works, but the buses/trucks still drive on the grass. I really need them to stay on the concrete. So like you, I experimented reducing the size of the parking spot. I found I can reduce the radius from 75ft to 55ft and still have a/c show up:

So now my buses/trucks drive round the edge of the dispersal pan. I have however found one wrinkle - if two adjacent dispersals need servicing, the bus/truck will not return along the original vehicle track but will take a short cut through the parking spot (and the a/c) to pick up the taxiway - and then we're back to the original problem ! This can largely be averted by having sufficient buses/trucks to service each dispersal and/or careful design of the vehicle track. Still working on that. But I'm getting there !

BTW I noted your comments about T numbers. Not sure how much effect they have on 4-engined tail draggers, but I will explore.

Cheers

 

[gfxpilot]

This is a really interesting issue. MS made the FSx format based of course on their knowledge with adaptations to airports and equipment using them since 1960 onward. Ill have a look at using my LOWI version and try to place (if i find one) a lancaster in there and see what I get as well. Ill let you know asap

 

[gfxpilot]

OK, here is a dirty experiment I tried

Added a taxiway in grass off the runway and flagged it as a military parking spot. (1)
(2)
Here what it looked like complied - not pretty but its for testing anyway
(3)I called the truck/bus - and (4) then (5) where it stopped

I dropped in some differing aircraft just for fun - beside the DC I used an ATR, ERJ190 and 737 - the truck reacted the same to all and in all cases followed the line and parked as the shots in 5. However, I then engaged idiot mode and added in a 747 just for the heck of it to see it does (6).

Ive D/L'd a spitfire and tried that (7) so in all it worked as Id expect

HOWEVER!!!!!

A)You see as I did that the "grass areas instead of nice apron will be grassed (apron) but you can make change to colours of lines etc as the yellow standard look out of place
(8)

B) Looking at the situation having flown the Spit and the Dc exact parking is going to need some careful smoke and mirrors because in WW2 there were no ground markers and exact parking places so its a case of course of devious thinking to show where the best possible place will be to stop the aircraft having a rear end shunt from the buss and fuel truck - Perhaps some carefully placed "chocks" on the ground where the wheels should be aimed for - this might simulate the yellow stop line used in modern airports. Even some discrete tools cases/drollies light give a nice effect and also act as the parking stop area. That at least will get the pilot in the right call zone.

Having flown both the taildraggers and then landed and taxi'd to this parking place without something really clear its a guesstimation if you hit the invisible green zone at the right angle and in far enough to prevent the bus hitting the pine end. I did try dropping 2 motorway cones where the wheels would best sit and that worked. Obviously red motorway cones were not in use 1939 - 1946


The issue of the trucks going off road. I've played for ages with my Lowi version here - basically they don't behave at all. They will follow the roadway until they make their final appraco to the aircraft rear end the they go off road by cutting a diagonal (look at image 4) One way to mask this might to have some apron acting as a path so when its complied the pilot sees roadway right to the stop point next to the aircraft. Not sure if that would help but smoke and mirror again.

The issue of the one bus servicing two or more aircraft - might have to play it safe and have 2 or more "buses" available - Ill try and get the set up going again to see if there is an optimum version this side and then let you know over the next few days

 

[dickb]

I see pretty much the same. I spent a happy few hours on Saturday trying similar stuff, but didn't have time to write it up. Like you, I noted that the truck doesn't stay on track. Apart from cutting the odd corner, it tends to stay on track until it reaches a point which offers a quicker short cut to its preferred location, which is usually a point in the 5 o/c position reference to the a/c, and then it hares off across the grass. (I say usually, because I did at least one test where it ended up at the 1 o/c position ! Although I have to say I can't recreate that scenario now.)

It's also difficult to predict which way the truck will be pointing when it stops. I've had them pointing in pretty much every direction. Which of course means they reverse in different directions before setting off. The only constant seems to be that the drive in, reverse, and drive out describe an equilateral triangle.

Of course, all this unpredictability could be deliberate, and is even desirable from a realism point of view. What isn't desirable is buses/trucks driving through the aircraft !

As mentioned, I'm really only concerned with AI a/c, which will park accurately and consistently. But you make a good point about parking user a/c without some kind of reference pointers.

 

[dickb]

This gets weirder and weirder. Check out this guy -

First pic shows the layout in ADE. Not pretty, just for test purposes.

#1 - matey sets off fine up the vehicle path. He then veers off right toward the taxiway and parking spot, but comes to a halt on the grass.
#2 - he then crawls at snail's base until he reaches the taxiway.
#3 - after another pause, he reverses thru 180 degrees onto the dispersal pan
#4 - he then drives forward through the parked a/c
#5 - on rejoining the vehicle track he turns left and drives away.

Go figure.