is 8 kt of crosswind enough to flip you over?
I don’t know that it’s enough to actually flip you over, but, yes, 8 knots is enough to make a very messy landing (and scary when you’re inexperienced) if you don’t fly the plane appropriately to counter it.
I dunno about that. Wind gusts I’ve experienced in real life seem to happen pretty instantaneously. And somehow always seem to be strongest when you’re bending over to pick something up off the floor that you just dropped. 
(not on final obviously, but when just blithely cruising along).
The weathervaning effect in the sim is overstated with regard to tricycle-gear aircraft as it doesn’t seem to account for friction and inertia. In all my years of flying, I’ve never had to use left rudder in a single-engine plane to maintain centerline.
I’ve had to use less right rudder, even closer to zero in the strongest crosswinds, but never left, and certainly not strongly left as I’ve had to several times in the sim.
Twin-engine is a slightly different story, but I think the bug is evident enough with the singles.
To clarify: this is in regard to takeoff in tricycle-gear aircraft.
3 Likes
Hmm, are you sure? As someone else noted here in one of these threads, they said they thought the same thing, until they went back and actually watched what they were doing, and realized that what they did with the rudder was so automatic, they didn’t realize what they were doing.
I can say I’ve definitely had to apply rudder based on the crosswind to maintain centerline during takeoff and landing.
What airplanes do you fly? (I’ve flown mostly Cessnas and Cherokees, and am low time even though I got my license in '96)
I can remember one time flying into Nantucket where I was crabbed nearly perpendicular to the runway on approach the cross-wind was so hard (I forget what it was, probably ~12 knots). I was the last plane they had land on 6 and then they switched to 15 right after me. Landing was fine tho. my wife/passenger was pretty, um, impressed shall we say. 
She did fly with me again many times
1 Like
I’ve got a lot of crosswind takeoffs and landings logged, and minor corrections aside, I can’t remember having to ever sustain left rudder. Maybe I’m doing it so subconsciously that I don’t realize it. My last flight in a 172 had a 9 knot direct right crosswind and I still had to use right rudder. Not a ton, but it was there - and I was paying attention because of what’s been going on in the sim. Heck, I have video of it, haha.
1 Like
Been there, done that as well, doing a precautionary landing ahead of a rapidly-closing front (I was flying toward it). Gusty as heck, airport on a hill, and a direct crosswind, so I was getting all sorts of loss on final. Yuck.
Sorry, don’t want to take this off-topic. The 310R should have some left-turning tendency, more than, say the PA44, because both engines on the 310 are conventional, but I’m still having a hard time with the feel of the weathervaning on a trike.
1 Like
Honestly, I’m going to have to think about which way I apply rudder for a given cross-wind. I’m working right now so don’t have time to really think about it, and, as noted, it’s so automatic I don’t pay attention, I just do it. I’m reading these notes and I trust what people are saying, and want to work it out for myself. I haven’t flown a crosswind landing in the sim in a very long time, so I can’t speak to the accuracy of the sim at the moment. I do know, however, that I know I use a lot of rudder and aileron in a strong crosswind to maintain centerline on both takeoff and landing… 7 knots being a decent crosswind to me, enough that I have to pay attention to what I’m doing. (I don’t tend to fly a lot on windy days). Too many other things to do than spend money on getting tossed about for the sake of hamburger in a cool place
1 Like
The key is pretty instantaneously. It’s “pretty” instantaneous because we’re using our perception as the timeframe for measurement. But nothing is truly instantaneous, as we know… The problem (I think) that RR is complaining about is because the simulator has a hard and fast frame for its calculations - the processor cycle used by the code guts crunching all the maths, whereas nature doesn’t (except for Planck time… but that’s another issue entirely. ) No matter how quick your processor runs, it’s still an incredibly coarse sample compared to reality’s trillions of data points per picosecond (as each atom has its own status that is “measured” each “cycle” to collectively come up with a wind gust). And what Robert is saying, I think, is that the data being input is either too coarse for the sim to process correctly, or not coarse enough for the sim to show gradations between the start and end point of your analysis.
To better illustrate: think of doing a jibe in a sailboat. Hull direction changes, sail pivots and ruffles and then catches the wind from the other direction and snaps to attention. In reality, it’s a smooth and relatively predicable process.
Now imagine simulating it. You’re going to have multiple variables at play – the wind force and direction, the angle of the hull, the current, the position of the boom and ropes, etc etc etc., all of which will get plopped into a bunch of code to determine what’s happening at a given moment. Here’s where the timeframe comes into play. Say the sail/boom position is updated every 1/100th of a second, but the relative wind data is updated every 5 seconds. (Relative because the boat is physically turning with respect to the wind.) If your real-life jibe took 3 seconds to complete, the sim would snap from one status to the other instantaneously, because it all would happen between data updates. You’d go from “boom on left” to “boom on right” in an instant, because that’s what the data is telling the sim. If your wind data was at 1/100th of a second, the changes would be reflected properly in the sim, and you’d have a gradual shift in the sail based on the actual change in the relative wind direction over the course of those 3 seconds. That would be a good sim! If the sim was working at, say, 1/10th of a second but the data was coming in at 1/100th, you’d get the correct shift, but it would be much jankier than it would be in real life, because the sim couldn’t process the nuances coming from the hard data. And so on and so forth. And this could also apply with respect to whatever process is calcuating that wind variable that’s getting sent to the sim. Simulation is hard!
To me (and, apparently, to Robert), it’s clear that there’s some kind of disconnect in the sim between the wind/turbulance data and the sim’s core engine, and probably also in the transition from ground to flight mode, that’s throwing the data processing off and resulting in these wild control swings being output. And sadly that’s beyond anyone’s ability to fix except for Asobo. I think PDMG just applies ex post facto fixes in their code – which means they probably have to redo all their fixes anytime there’s some small change in the engine or the wind calculation structure (which is probably why he’s so annoyed with them 
), but that obviously doesn’t change anything with any of the other aircraft, who are still subject to the flawed base. Sigh. Maybe someday…
To keep this relatively on topic: the 310 is an awesome plane and everyone should buy it.
3 Likes
I think you can select modifiers. I’ll check next time I’m on.
With regard to the off-topic crosswind discussion - in real life you do not consciously think “I better apply left (or right) rudder”. You just keep the nose straight with rudder and often do not even notice which way you needed to move it to adjust.
1 Like
I’m pretty sure, as the developers have said, that the problem is in the transition code. They’ve discussed how
- The tire friction model is completely wonked. They have done some work on this, I’m not sure the technical details other than, until I think SU9, it was the FSX point based code. They have said they’ve added an improved friction model now, but I’m pretty sure they’re only partially done (which, as we know, usually makes things worse).
- Also from FSX, there was an instantaneous change from tire friction in action and no side force (wind didn’t affect planes when they were on the ground except in the direction of travel), to suddenly experiencing the wind from all angles instantaneously as soon as the wheels were deemed off the ground. I believe they have done some work on having wind affect planes from the side when they are moving but still on the ground, I’m not sure if they are done yet.
I would be really confused if the effect of wind or the gusts on the plane were calculated at a different frequency from the time of calculation of all the forces on the plane. While what you said makes sense to me as being a problem, the concept doesn’t make any sense to me as a method of calculation on applying forces to the plane (the out of sync part). Now, maybe there’s an issue that gusts are too strong from one calculation to the next, I could see that; the mitigation being to, since you know the gust is coming, increase it a little bit at time over each force calculation from zero to full force over say 1 second or some other better interval.
I do believe there are now tools in developer mode that allow you to see the affect of the air flow on aircraft to confirm all this.
1 Like
do we really have to label every subtopic developing out of us flying the 310 in different scenarios - and sharing our questions and experiences, then discussing a bit the underlying physics and modelling of it - as “off topic”. Makes this thread a bit sterile, should it really only be about “where is the update” and “love it, yeah”?
IMO it stifles an enlightening discussion, if occasional sidesteps to discussing underlying fundamentals are forbidden.
How does it work in MSFS?
Basically they have two options.
a) sampling, using sampled data from the environment at a frequency/sample rate so high (e.g. thousands of cycles per second) that the jagged steps of changing values are smoothed out by their small proportion over the very short time span of one sample period. (I guess the source data is always coarse at long sampling intervals, so interpolation would be necessary before injecting the data)
b) vectoring, using sampled data at coarse intervals, e.g. once every few seconds or worse, then using mathematical models to gradually change the data without immediate step changes to arrive at a subsequent sample gradually over time. Basically interpolation between coarse data points in time.
It seems they are doing it simplistically, kind of combining the worst of both methods? Sampling at coarse intervals, then injecting data more or less raw into the calculations without interpolation math?
And how does the gust logic work mathematically? it seems also rougher, more sudden, than IRL to me. Obviously gusts are 100% synthetic, since there is no real world measurement system, that measure and reports gusts in real time at a high enough sampling rate.
How often are Live Weather parameters sampled?
On Topic - is the 310 crosswind handling modelled correctly in take-off ?
Off Topic - let us chat about basic aeronautical knowledge about crosswind takeoffs
2 Likes
let us chat about basic aeronautical knowledge about crosswind takeoffs to evaluate If the 310 crosswind handling is modelled correctly in take-off ?
on-topic or off-topic?
1 Like
Can anybody tell me if this plane ever goes on sale or is it always the same price? I’m thinking this will be my next purchase but I’m in no hurry for another add-on.
It shouldn’t it’s excellent! Great bang for your buck.
3 Likes
We rarely, if ever, do sales.
1 Like
Hej people of the internet 
first i love the 310 its my all-time favorit. But after the last update i cant pop out the gns530 window to my small gns530 screen there is only an black window. The Asobo gns530 doesn’t work like the pms50 gns530. In another airplanes like the piper arrow 3/4 it works fine also the 420 in the 310. I have no other addons only navigraph, onair manager and the piper turbo arrow. The 310 is new installed also the 530 and msfs… nothing helps. is there an bug in the 310 itself?
Moe❤️
Well I bit the bullet and bought the 310 and I must admit I am very impressed!
A couple of initial impressions:
- Is it really so reactive to control inputs? I am used to flying stable (trainer) aircraft and this one seems to control more like an aerobatic plane! It can barrel roll easily which I was not expecting. Is this right for the C310R??
- The yoke animation doesn’t match my control input. About 1/2 aileron deflection on my yoke shows in the sim as 100% aileron deflection on the yoke (but looking at the wings the actual control surfaces only get to 100% when I turn my yoke all the way). Is this a bug with the add-on?
- Power seems low on one engine. I killed the left engine (and feathered it) and found it was only just possible to maintain altitude with the right engine full throttle and full fine prop. This was with only 2 people on board and 50% fuel (and clean airframe) so I can imagine fully loaded this would be a scary situation! Is this expected?
- EDIT: Also I noticed my control bindings for increase / decrease cowl flaps didn’t seem to work.
I am yet to look though the user guide properly so I might be missing things but I wanted to have a play with it first before digging into the details. These were just the first things that stood out to me after an hour or so of mucking around.
1 Like