HWasp

"So the procedure goes as follows: 280 kt indicated should give you 14 nm for 10 000 ft. 280 is chosen because it's close to the best glide speed (~260 kt) and inside the engine relight domain. The slats should be forced in. Stabilize at 5000 ft to slow down to 230 kt minimum. Maintain 230 kt in final. This procedure does not give fuel or configuration information but we can guess that in such emergency external stores would be jettisoned so the procedure must be possible with basically full fuel and no stores. I'll record a track later." This was the procedure given to me. I was testing this one.

For this I have also disabled the slats, and made an S turn to simulate a 30 degree offset, so worst case scenario. I would consider this reasonably safe in this config

Yes I did, but I have found something more interesting: I just stumbled upon this combination : with 1 full center tank and 100% fuel, sustained rate at 340 kts is almost exactly 4,8G as per the video (that may or may not be relevant) Also with the same configuration engine out procedure is fully safe (the previous loadout was really on the limit) M2000_engine_out_proc_1droptank_1122.trk

Thanks! I checked that, really is on the limit as you say. I would not have thought, that the aircraft is capable of this (I don't think I ever flew that high with less than Mach 1,5 ) On the other hand, regarding the possible effects of a 10% L/D reduction, I did fly the engine out profile now with full A/A loadout + a droptank at full fuel, and still made it. I have even made a small S turn around 5000 feet to simulate the effect of a 30 degrees offset. I'm not sure, what the combined drag penalty of the 2x Magic2 + 2x 530D + droptank is, but I would guess it is a lot. Please see the track attached. I understand, that emergency procedures have tolerances built in, but this seems too much for me. Speed can be completely off of the optimum speed (demonstrated earlier), stores, droptanks can still be on, even with a turn. M2000_engine_out_proc_err_FullAAload_droptank_1122.trk

Could you please tell me a possible FL500 loadout? Thanks! Regarding the chevrons: I think they don't follow the changes perfectly intantenously in the HUD footage. There seems to be a maximum speed, at which they move, so during a high G-onset rate they are a bit late. Looking carefully at the video you can see at high G-onset turns, that the speed is already decreasing while the chevrons still indicate an increase. That was my main point, why the initially claimed G at Ps0 is not valid, because the chevrons are late and plane is already beyond Ps0 when they arrive at the FPM.

I'm not ingoring anything, I brought up points that might prove otherwise.

I understand, it is complicated. Are you sure, a Cd increase causing a 0,5-0,7 G sustained G drop at around 12 AoA would distort that much? Again I'm thinking about relatively minor differences.

No, I gave you some numbers, that seem to line up. Is Jx value showing acceleration? Is the -5 engine is weaker, by 10%? How does a weaker engine affect acceleration? Does temperature affect thrust? If -5 has 10% less thrust than P2, at 15C, Jx will be noticably lower. You said 15C difference causes around 5% thrust difference for the engine--->> need a LOT of temperature correction to make up for the 10%, (-15C just going by your numbers ) I simply just used the information You have given me, and they seem plausible to me.

M53-5 has 8-10% less thrust than P2, right? This means acceleration will be less if T=15C, so Jx< 0.68 Previously you have mentioned, for this engine 15C temp increase is roughly -5% thrust That means, that if you want to make up for the difference with temperature, then this should be a winter video, with temps around 0 Celsius at least. So actually with that in mind, since foliage is visible in the vid, and it does not look like winter to me, this could still be a C model with non-standard B/W camera... This would be best case scenario for everyone, since even though it means the necessity of another round of tuning, but finally there would be some really solid evidence for STR, and also probably this heavily debated subject about STR could be put to rest, and you guys could focus on my new MiG-23

That is certainly wrong, as this type of airbrake does disturb the airflow and may be even altering vortex forms and generation, so the whole system might act differently compared to clean. Also, this is not even a precision manuever, it's just an overhead break. All kinds of variables there, great for testing against gross errors, but otherwise...

Also, even though I understand now, what Kercheiz said about the signs, that show, it is a B, and while I trust his knowledge of course, that is not 100% evidence, although it could be very likely. if Jx=0,68, what are the chances, that temperature there is just right for the B to match the C's acceleration exactly. Of course it can happen, but still...

Ok, that is understandable, but given the comlete lack of data otherwise, it might still be better to assume, that it has similiar thrust profile, and use that, (I really wouldn't think, that a slightly improved version of the same, with the same intake, etc, would be that completely different...) Again, I really can't see how the break procedure can help us with this +/-0,5-0,7G difference. Especially with the airbrakes involved. It works, of course, but my problem is with drag between 5 - 12-ish AoA clean. I can't test that with airbrakes out at all....

I have on more point for the HUD tape video: The Jx=0,68 on take-off. If that matches exactly the value we need on take-off on this version, that means in the vid external conditions (lower temperature) compensate for the different engine. If we could agree, that because Jx=0,68, ---> temperature is considerably lower than 15C ---> thrust in those conditions does actually match our version, because of the matching acceleration, that would mean, we only have to account for the possible weight difference of the B version (I know there are aerodynamic differences, but given the total lack of usable data...)

That is a very big claim, given there is no info there about actual thrust. Given, there are that many unknowns, it would be still better to try and interpolate data from the B-version video, because at least weight and thrust differences are known, so with those accounted for it might be less wrong, than vids like this. The previous FM might have had too much drag increase above 14-15 degrees AoA, certainly possible, but please keep in mind, I am talking about the 5-12 AoA range mainly. I think the high AoA range is ok, since it was fixed, and increasing the rate of drag increase slightly between 5-14 does not have to affect that area seriously.

Yes, and also to stay reasonably close to the best glide speed and not end up with a whole lot of induced drag at low speed range of the drag curve. Look, I can fly the whole thing between 170 and 180 kts easily. (And still have hydraulics in DCS, but this is a different subject then) The video looks like a partial engine failure, someone in the description says it was a control unit malfunction, so it's not a complete loss off thrust. How is that relevant for glide ratio? Here is the same slowing below 180 already at 10000 M2000_engine_out_proc_err_180_alltheway_1122.trk

I think I have found a problem here, testing the procedure: At 5000 feet I made an intentional error, and slowed down below 170 kts in level flight instead of the 230 minimum. Problem is, I still made it easily, actually had to drop the gear at 2000 feet. Also the flare is not a problem at all starting from this lower speed. 230 kts on final is a very high approach speed, there must be a good reason to keep it there. If I can do the same procedure with such an error, something must be off a bit. I tested the glide ratios at speeds 240 - 160: 240 : 8,1 AoA 5,5 220: 7,4 AoA 7 200: 6,75 AoA 8 180: 6 AoA 10 160: 5,4 AoA 12 I'm sure the approx 8 L/D base given by this procedure must be ok, but I find it very strange, that plane could still have 5+ L/D at 160 kts where AoA is 12 I would be very interested to see, what would happen, if the drag curve would be steepened a bit to hit the 4,8G / 340 sustained according to the video (I think AoA around 12) Please see track attached: M2000_engine_out_proc_err_170_1122.trk

Ok, thanks. It works, tried it, but I think this test is a bit "low resulotion" regarding the +/- 0,5 or 0,7 Gs sustained at 340, we are looking for. This is a very good reality check for gross errors, but I don't think there is any. I would call this whole discussion "performance tuning discussion", and for this matter a HUD video, like the one you posted, or things with similiar detail are suitable I think. It's very unfortunate, that hud footage is a 2 seater with different engine.

Thanks! Is this a straight in procedure, or does it start over the airfield similiar to the MiG-21 for example? This would be quite important, as turns increase descent rate quite a lot.

Thanks, now I understand better. That would be great, sounds interesting!

To be honest, I did not fully understand the points Kercheiz brought, on how exactly are we sure about what version this aircraft is. Information is very scarce, it's quite sad, if one of the only sources is not valid. While I do really like the module and the FM overall now, it would be really important to back this up with some solid sources, because otherwise it will be constant debate without an end.

For me this data point shows around +10%+ overperformance in STR at that speed.

Considering the 4,8G at 340 kts in the vid at around +1000 feet/minute (Ps= +16,6 feet/sec), I think the realistic STR G based on the video is around 4,9 to 5 Gs. +16,6 feet/second is really not that much looking at other EM diagrams, so I would consider this much less of a problem compared to the chevron being late at a high g-onset rate. The chart would look like this:

Ok, thanks, but how much is the difference? The screen from the video was shown by myHelljumper to validate the current STR. My point, to summarize is: When the pilot increases the G-load at a high rate (as it is the case with the 02:15 screenshot), the chevrons are lagging behind slightly, and at the point the screenshot was made, we are already beyond the G load for sustained turn at that speed. We can see that happen in the other direction as well, when the pilot unloads at 02:13 and G is only 4,8 while chevrons indicate Ps0. That would also be an incorrect data point to the opposite direction. The only moment I found in the video, where speed is reasonably stable for some time in a close to horizontal turn, is the turn at 02:03. Please consider this turn as a data point instead of the one shown previously, even though there is a around 1000 feet per minute climb there, this might be the most accurate. How on Earth would a HUD footage with speed, g load, etc shown not be usable data??? This is usable, we just need to be careful about certain things, like lagging indicators.

One further point against the chevrons for me is the pull at 02:10 Please see the screens: At 02:10 Speed is 409 and chevrons are indicating a large increase At 02:11 Speed is already dropped to 406, but the chevrons have just only reached the FPM This tells me, that there is a little delay in the indication, so I think there is a possibility of error while using the chevrons to determine the moment where PS=0. In this case, because your data point is at a moment, where G load is being increased quickly, I think at the moment where the chevrons indicate Ps0, we are already beyond that in the video.

Another interesting data point for me is at 02:03, where the pilot performs the only almost horizontal sustained turn for a short time, so speed is stable there actually, and not just going through the range quickly. There, with a very subtle climb rate, G sustained is 4,7-4,8G I tried to replicate this as close as I could in DCS, with 86% fuel load (I figured, 2 minutes of AB would get me there at that point), and DCS value is different, overperforming a bit. So for me there is 5,4 G vs 4,8 at a slightly lower speed and a very little bit higher V/S For me this 02:03 moment is the closest thing to an actual sustained turn, even though it's quite short, but speed is finally stable for a bit, and not just going through the range in a single screen