fixed S3B tanker does not have enough fuel, or doesn't dispense enough fuel before ending the refueling task

[Rolds]

To test this I created a situation where I had a hornet with two fuel tanks and dumped down to ~2,300 lbs and then tanked to see how long the S3 would dispense. See track. When my fuel showed 4,900 lbs, the S3 retracted the drogue. I could no longer contact it on radio to begin refueling again, I believe it had hit its bingo point and considered its refueling task finished.

The tanker began this mission in the air with a full load, as did I, so very little fuel would have been used up prior to me tanking. In the mission editor I see that full of fuel the S3 weighs 38,775 lbs, and empty it weighs 26,649lbs, giving fuel weight of 12,126 lbs. The tanker gave me 2,600lbs, maybe burned 1,000 though I doubt it would have been that much in this short period of time (just watch the track), so I think it stopped refueling with at least 8,500 lbs and the carrier it is set to land at only 30 miles away. 

Other experiences with the S3 tanker had made me suspicious as I had seen it stop its refueling task mid refuel on many other occasions. I suspect there is a math error in the code somewhere, like the outflow rate is measured in the wrong units, or not all the internal fuel is usable, or its bingo logic is far too conservative.

S3B does not dispense enough fuel.trk

[Flappie]

Hi Rolds, a fix was applied with the latest OB. Can you please recheck?

[Northstar98]

29 minutes ago, Flappie said:

Hi Rolds, a fix was applied with the latest OB. Can you please recheck?

There's been a change compared to 2.9.1, but there's still some funny business going on with how much fuel the S-3B tanker has available and how much it weighs. Though haven't tested how much fuel it dispenses.

[Exorcet]

It seems to make sense with my testing:

S3Fuel.trk

[Exorcet]

@Flappie @Rolds

 

While the fuel transferred seem right, the issue persists. It is a bingo fuel issue. If the S-3 is too far from an airbase, it holds too much for bingo:

Here the Tomcat is forced to abort because the S-3 suddenly ends refuel task.

 

It's because I moved everything out here:

 

 

 

S3Fuel2.trk

EDIT - I went back and added a carrier under the S-3's route and this seemed to fix the issue, unlike in the original case. So maybe it is actually working as intended.

 

EDIT 2 - I converted the original track to a mission and added a fuel script, turned on RTB bingo and it looks like the S-3 gave as much as it could, so I guess it's fixed now.

Edited December 24, 2023 by Exorcet

[Rolds]

4 hours ago, Flappie said:

Hi Rolds, a fix was applied with the latest OB. Can you please recheck?

Hi Flappie,

EDIT - THERE IS STILL SOMETHING WRONG SEE NEXT POST

I am re-running the test. I notice a couple of things before I start. The S-3 can now carry 15,183lbs of fuel and its weight when full of fuel is 41,833lbs per the mission editor, an increase over the  12,126lbs fuel weight and 38,775 lbs total aircraft weight I noted in my original post. I also noticed that the tanker now starts with 80% fuel whereas before I had it set to 100%. Looking at other missions I made prior to the most recent patch, it appears all S3 tankers are now starting in the air with 80% but I almost certainly placed them as air starts with 100% fuel. People who want to maximize the tankers' fuel load might want to recheck old missions.

Re-running the original flight, I dumped to 2,400 then tanked. The tanker refueled me to 11,260 then retracted the drogue, so this time I think it gave me about 8,860lbs. That is a definite improvement over the original situation. This may mean the bug is squashed, but my method of testing is crude compared to @Exorcet script. I looked at that script and it is beyond me but that looks like the right tool to test this.

Edited December 24, 2023 by Rolds
update

[Rolds]

I re-ran the test with @Exorcet script in place. In this test F18 fuel at the start of tanking was 3,177 lbs and at the end was 11,676lbs for an increase of 8,499 lbs. In all cases the script output agreed with the hornet in cockpit fuel gauge very closely.

At the start of refueling the S3 fuel from the script read 14,105lbs and at the end it read 1,048lbs for a decrease of 13,057 lbs. In other words the S3 lost 4,558 lbs more than it gave the hornet, and it is not reasonable that it burned that much fuel itself during the refueling. The S3 then returned to the carrier, did a case1 and landed with only 252lbs remaining.

Of course there is the possibility the script is wrong but I don't think so. To test this, I had the S3 airstart with 3% internal fuel and watched it, and sure enough when fuel reached zero according the the script, the crew ejected.

I think there is some problem where the fuel outflow rate from the S3 is higher than the rate it is going into the tanking aircraft, and the difference is just being lost.

EDIT - Attached is a track you can play around with which includes the fuel script. In this case tanking began with 2,806lbs in the hornet and 14,290lbs in the S3. It ended with 11,330lbs in the hornet and 1,144lbs in the S3. The hornet gained 8,524lbs while the S3 lost 13,146lbs.

S3B fuel rate issue.trk

Edited December 24, 2023 by Rolds
add track

[gulredrel]

Glad I found this and it's something known. Got only around 4000lbs from the S-3 to my Tomcat and than the S-3 returned to the boat. Will need to check fuel level in ME.

On 12/24/2023 at 7:43 PM, Exorcet said:

added a carrier under the S-3's route

My S-3 launched from a carrier, so there should be no bingo long range problem I guess? 

[Exorcet]

20 hours ago, gulredrel said:

Glad I found this and it's something known. Got only around 4000lbs from the S-3 to my Tomcat and than the S-3 returned to the boat. Will need to check fuel level in ME.

My S-3 launched from a carrier, so there should be no bingo long range problem I guess? 

If the carrier is nearby, then bingo shouldn't be an issue, although unrelated to all of this I did observe some F-18's refusing to land on a carrier in one of the my missions the other day. I did not look into it very deeply, but I wonder if there is some issue with AI failing to recognize the carrier as a landing spot sometimes.

[Flappie]

On 12/24/2023 at 10:11 PM, Rolds said:

I re-ran the test with @Exorcet script in place. In this test F18 fuel at the start of tanking was 3,177 lbs and at the end was 11,676lbs for an increase of 8,499 lbs. In all cases the script output agreed with the hornet in cockpit fuel gauge very closely.

At the start of refueling the S3 fuel from the script read 14,105lbs and at the end it read 1,048lbs for a decrease of 13,057 lbs. In other words the S3 lost 4,558 lbs more than it gave the hornet, and it is not reasonable that it burned that much fuel itself during the refueling. The S3 then returned to the carrier, did a case1 and landed with only 252lbs remaining.

Of course there is the possibility the script is wrong but I don't think so. To test this, I had the S3 airstart with 3% internal fuel and watched it, and sure enough when fuel reached zero according the the script, the crew ejected.

I think there is some problem where the fuel outflow rate from the S3 is higher than the rate it is going into the tanking aircraft, and the difference is just being lost.

EDIT - Attached is a track you can play around with which includes the fuel script. In this case tanking began with 2,806lbs in the hornet and 14,290lbs in the S3. It ended with 11,330lbs in the hornet and 1,144lbs in the S3. The hornet gained 8,524lbs while the S3 lost 13,146lbs.

S3B fuel rate issue.trk 2.89 MB · 0 downloads

 

Hi @Rolds. Sorry for being so late, I was very busy. Your track is just great. I'm able to determine that, at a speed of 322 knots:

• The unconnected S-3 loses ~94 lbs / min.
• The unconnected F/A-18C loses ~69 lbs / min.
• The connected S-3 loses ~4492 lbs / min.
• The connected F/A-18C gets ~2884 lbs / min.

It means the S-3 gives ~4398 lbs/min while the F/A-18C receives... only ~2815lbs/min. The hose has a serious leak: 36% of the fuel is wasted!

From your calculations, I find the leak amounts 35%, so it seems we both got the figures correct.

Issue reported. Thanks for the hard detective work.

[Flappie]

@Rolds Devs ask for a Caucasus track. I tried my best to make one but I suck hard at AAR. Could you please make one? Here's the miz.

 

EDIT: Nevermind. A fix is in the works.

Caucasus_AI_S3B fuel rate issue.trk.miz

Edited April 1, 2024 by Flappie

[Rolds]

Your math may overstate the "leak rate". On S3 math, I agree.

I think the figures should be added for the F18C to determine how much it gains. Connected fuel gain rate=transfer rate less burn rate, therefore transfer rate=burn rate plus fuel gain (which is better thought of as net fuel gain rate).

I imagine this won't matter to the coders. The issue is probably going to be something simple like the transfer rates are hard-coded for each aircraft and independent of the aircraft type on the other side of the transfer. Simple to understand I mean, that may be difficult to solve.

 

[Flappie]

I believe this was fixed a few updates ago.