Saxman

Update: We're close to having the Zero dialed in. I've got it to the point that the Corsair can now extend away from the Zero as it should, and AI Corsairs can actually shoot them down. We've fixed quite a few issues so far: Engine displacement was too high. Corrected the gearbox ratio. Propeller diameter was too small. Manifold pressures were too high (it was pulling 67in MAP at WEP!). Removed ADI (a system that to my knowledge was only ever installed on the A6M6, and never actually saw combat). Sea Level speed was too high (there's some debate over whether this actually affects AI performance or not, but this was the real key for me when I tested it). Roll rate has been corrected to tested values (max roll rate is 56deg/s at 150mph tas and falling from there). We're still looking at the other aerodynamic data. If anyone has info for the A6M5a or could help with calculations it would be a big help.

My problem is that the pilot's left arm blocks your view of the trim tab indicators so you can't actually see how much trim you're using. Less of an issue when you're in the air and can judge it by the plane's behavior, but a serious problem when you're preparing to take off.

I think it's absolutely a major factor. There's a YouTube channel I watch regularly, and EVERY time he flies a Mustang or P-47 I see the same sequence of events: Pulls up well inside convergence range, often even closer than 100 yards. Makes a snapshot. A couple rounds plink off the target's wing on either side of the fuselage. Complains about .50cal being "underpowered." His response when someone else suggested he's firing out of convergence was an indignant, "I used my rudder to walk the rounds across the target," which doesn't address the issue, and if anything only makes it worse by scattering his hits even further.

So is this something for a future development? A leftover from early planning? It's odd to see them defined at all since the 1D never had the wing tanks.

The real important consideration here is you were shooting in convergence. To be fair, those Japanese tanks were basically made of tissue paper. Shermans were doing to them what Tigers were doing to Shermans.

It's possible the original mod was created before the I-16 was added.

People always complain about .50cal lethality, when it's really about not knowing how wing-mounted .50cal work. Most important of all: Make sure you're shooting in convergence range. This is actually easy to do with the Corsair: The Mark 8 sight reticle is calibrated so you can use the middle and outer rings to range a target if you know its wingspan. For a 30-foot wingspan (approximately the size of a single-engine fighter) the target will be at 400 yards when his wingtips are touching opposite sides of the 50mil (middle) ring. At 200 yards his wingtips will be touching the 100mil (outer) ring. Obviously if you're not dead six you're going to need to give it a bit of windage, but once you get used to it it's not too difficult. I THINK during the Gunnery Training the Corsair's guns are centered at 300m (330+ yards).* The 190's wingspan is 34ft. So you'll be at convergence range when his wingtips reach half way between the middle and outer rings. To simplify things, shooting him anywhere between 200-400 yards is close enough for hand grenades. The A6M5 wingspan is about 36ft, so you can comfortably use the same estimate (any single-engine WW2 fighter will be "close enough" to 30ft span you can use the same range estimation). CONVERGENCE MATTERS. The Browning.50cal is basically just throwing lumps of metal at the target. That's it. It's reliant almost entirely on kinetic impact to cause damage.** If you're firing outside your convergence range all you're going to do is scatter your fire across the target. Unless you have a golden BB that hits something important just right, you're just poking holes. To cause REAL damage you need to concentrate a whole bunch of BBs into a small point to amplify the energy of the impact. IE a single .50cal round won't break a wing spar. But putting a couple dozen into the same spot on it all at once is another matter. This applies both if you're too far away and too close. So if your sight picture looks like this: You're TOO DAMN CLOSE, and none of your bullets are going to be hitting in the same spot (they're going to miss center of pass entirely and go up the wings. You can best see how it works here: As you can see, there's a range at which you have your tightest concentration of fire. That's where the damage is done. You can change your convergence range when setting your loadout.* The second, is to be aware of your combat mix. The Corsair has two different mixes of ammunition for air-to-air: Combat Mix and Combat Mix Late. These mixes are grouped in five round patterns. I believe Combat Mix consists of 2 ball rounds, an armor-piercing round, an incendiary round, and a tracer. Each of these types of munitions have different properties. Late Mix is four Armor Piercing Incendiaries, followed by a tracer. Late Mix is going to cause considerably more damage. Finally, remember that the Fw-190A is a considerably more ruggedly-built fighter than the Zero. It has good armor for the pilot and other critical systems, as well as self-sealing fuel tanks. This is, naturally, going to be much more resilient than the Zero which has none of this. However, a good one-second burst in convergence range, especially with the Late Mix, will really mess him up. * - Currently you can set convergence at ranges between 300 - 500 meters. THIS IS INCORRECT, and I've already started a thread on the wishlist to get it changed. It is on average at considerably longer distances than the US centered their guns, which tended to be between 200-400 yards (yards, not meters!). However, we also have accounts of some Marine squadrons setting their guns as close as 100 yards! PLEASE, M3, can we get these ranges adjusted? ** It's a bit more complicated than that. You have ball, which is just an inert lump of metal. Armor-piercing are hardened inert lumps of metal designed to punch through armor. Tracers are nominally inert lumps of metal trailing an incendiary component that makes them visible and can, under the right circumstances, start a fire. Incendiaries have a small charge designed to light things on fire, though their explosive power isn't remotely comparable to a cannon shell. Finally, you have API, or armor-piercing incendiaries, which combine the properties of AP and Incendiary into one package. By the end of the War, the US was almost entirely using API for air-to-air combat loads.

I think we've made a breakthrough: The Zero copied aerodynamic data from the P-51 for the SFM, but the original mod-makers did NOT change the engine MAP. Which means at full power it was pulling more than 50% MAP over what it should (the Sakae 21 engine should max out at about 42in MAP. It was pulling 67!)

So I made an interesting discovery yesterday. I was poking around through the luas to possibly mod the 1D to create a 1A, and much to my surprise I found this reference in the damage tables: The left and right wing tanks are already defined in the damage table, but are commented out. I haven't dug into it yet to see if the tanks are established anywhere else, but it makes me wonder if M3 is already planning to add additional variants beyond the Corsair Mk.IV.

Not really. Before and at the beginning of the War, they were formatted as Squadron - Mission - Plane Number. IE, Plane #2 of VF-3 would be marked 3-F-2. A plane of Scouting 6 would be marked 6-S-#, etc. By mid-1942 the squadron number was occasionally omitted. So you'd see F-2, S-6, T-15, or B-4, etc. Marine squadrons used the same system, just adding an M in front of the mission. So, 212-MF-5 for aircraft #5 of Marine Fighting 212 (Marine Air Groups were nominally designated as the first two digits being the Air Group number, the third digit being the squadron number. So VMF-212 was squadron #2 of MAG-21. In practice this wasn't adhered to due to the necessity of squadrons being moved where they were needed, not necessarily to keep their air groups together). The squadron number sometimes got omitted from the Marines' machines as well. IE Bob Galer's Wildcat on Guadalcanal was marked MF1. By the time of Guadalcanal this was starting to get phased out, and the squadron and mission were being omitted, leaving only the plane number. The squadron and mission would no longer be indicated in the MoDex number. I believe this was at least partially to help with operational security (the same reason the Navy stopped matching squadron numbers to the carriers' hull numbers: It made it harder for Japanese intelligence to determine what carriers were where based on what squadrons were operating). The full number carried on a little into 1943 (VF-17's Corsairs were clearly marked 17-F-# during their carrier trials in April). However by mid-1943 the system was gone entirely. At this point you only had the plane number. For Marines, this was usually derived from the last 3 digits of the serial number, as in #883 and #740 of VMF-214 (17883 and 17740, respectively). The Navy simply assigned a number to the air frame. The only time you started to see carrier-specific markings come into play again was late in the War when you had large-scale operations involving multiple carriers and flight controllers needed to organize the chaos (IE Bunker Hill's white arrow).

I'm aware of this. However there's been some reports that the engine may be modeled as the -8 version of the engine, rather than the 8W (water injected). So I was basically trying whatever to see what stuck. The water injection button turns the water injection system on. This is one of the questions about the current engine model because injection in the Corsair was automatic based on throttle position, not a separate control. Which is about 10mph too slow. At critical altitude the F4U-1D should be about 410mph TAS. There's not a speed table for the 1D that I'm aware of, all of the tests we have data for are for seem to be from F4U-1As. The 1D loses an average of 10mph over the 1A because of drag for the knuckle pylons, so it can be estimated from there.

I'll stick to my USN/USMC MoDex numbers.

The odd thing is I can (almost) never get full MAP with injection on. The needle almost never gets above 54in. And when I can I can never figure out what combinations of settings get it there. It seems sometimes going to Auto Rich boots it up, sometimes going into high blower even below critical altitude, sometimes NO blower.

Does anyone have any good aerodynamics data for the A6M5a? I'm going to try to fix the flight modeling so it behaves a bit more realistically, but so far all I have is rate of roll data (peaks at about 56deg/s at 150mph, and goes down from there. Dol's numbers were WAY off) and some VERY basic calculations for thrust based on airspeed and horsepower. The specific info I'm looking for: -- Cx0 - Coefficient, drag, profile, of the airplane -- Cya - Normal force coefficient of the wing and body of the aircraft in the normal direction to that of flight. Inversely proportional to the available G-loading at any Mach value. (lower the Cya value, higher G available) per 1 degree AOA -- B - Polar quad coeff -- B4 - Polar 4th power coeff -- Aldop - Alfadop Max AOA at current M - departure threshold -- Cymax - Coefficient, lift, maximum possible (ignores other calculations if current Cy > Cymax)

Check your trim, beyond that there's definitely some odd behavior with the speed that's been reported several times. Even with injection on it's hit or miss whether the engine ever gets about 54in of manifold pressure.

You can't outrun them. You can't outclimb them. You can't outdive them. You can't outroll them. Even the F4UFO AI can't fight them. Seriously, the flight modeling is completely borked. At least the 190s you can actually get guns on.

I think Dol's Zero is the main one, but its flight modeling is...suspect.

What really annoys me is not only can Dol's Zeke outrun a Corsair in the dive, it can CONTINUE MANEUVERING. The Zero shouldn't be able to roll at all above 300mph because the stick basically freezes from the control forces.

Good like diving away. Especially if you're running Dol's Zero.

Ram had to do with air flow to the carburetor: Ram air was the same as Direct. Alternate took heated air from the engine cowling.

I always feel that no matter how much speed I've got the Corsair simply dies in the vertical. And unfortunately, the AI LOVES going vertical (and can climb like a rocket). Although when I can get guns on target that Late War mix sure has no trouble lighting them up. Although honestly, there's times I've found AI wingmen all but useless.

MP changes seem to be hit or miss. Sometimes I'll see an increase in MP, others not. There's also times where the MP doesn't change until I shift the Supercharger, even if I'm below critical altitude for that gear.

Keep in mind about 50gal of that 237 is the reserve.

So tonight I went up earlier to do some testing on the blow-up flaps, but also spent some time doing some speed testing. At 22,000ft, trimmed for level flight, cowl flaps, oil cooler, and supercharger closed, full throttle, WEP switch to "on." I managed to JUST get to 250kts IAS. With a surface temperature of 28 degrees C and a calculated -15 degrees at altitude, that gives me a TAS of 364kts, about 418mph. That's a little faster than the F4U-1D is supposed to be. I didn't notice any real difference in speed between Auto Lean and Auto Rich.

Track uploaded to Dropbox here. There's also some general speed test stuff at the beginning. The flaps are at the end. Dive from about 20,000ft with flaps on the 20 degree (2nd) notch. Airspeed cleared 400kts on the dive, and at the end I still had to manually retract the flaps (it's possible it might have blown up from 20 degrees to 10 degrees on its own, (or I just missed the second notch and only had 10 degrees from the start) but I DEFINITELY had to retract the 10 degree notch manually, which should not have been the case.