Gimbal distance and Speed Range Estimates using Lines of Bearing and/or DCS

8190E22A-9C5A-4C31-A9F3-1701C2DA6F2D.jpegVenus can be seen in the daytime for sure if you know exactly where to look. You don’t need a telescope.

I took this photo of Venus at 11:43am on January 1, 2019 with a Canon D77and my zoom lens at 300mm (iso 100, f7.1, 1/500sec)

the brightness of Venus is higher than the crescent moon. I saw the two earlier morning when the sky was dark and that helped me find it later in the day and take this shot.
 
8190E22A-9C5A-4C31-A9F3-1701C2DA6F2D.jpegVenus can be seen in the daytime for sure if you know exactly where to look. You don’t need a telescope.

I took this photo of Venus at 11:43am on January 1, 2019 with a Canon D77and my zoom lens at 300mm (iso 100, f7.1, 1/500sec)

the brightness of Venus is higher than the crescent moon. I saw the two earlier morning when the sky was dark and that helped me find it later in the day and take this shot.

When you took that (really nice) photo, Venus was at the edge of its orbit, when it is the brightest, due to its phase and distance to the Earth (0.64 AU, being 1 AU the distance Earth - Sun)
stellarium-004.png


For the gimbal video, AFAIK, it was some day in January 2015. By that time, Venus was at the far side of the orbit, "behind" the sun. Even if its phase was full, it was a 1.5 AU of distance. Also, as the gimbal object is near the horizon, it means it had to be either before 9 am in the morning (with the sun already up in the sky, making it more difficult to see venus), or near 7 pm, when the sun already set, it's darker and it may be easier to spot venus in IR (exact hours depends on the time zone and place, but anyway, it's just a quick look.)
stellarium-002.png
stellarium-003.png

Anyway, the apparent size of Venus is so small that even if it could be seen in MWIR (which I doubt), in a 640 pixel FPA with 0.7º FoV it would be as big as ...2 pixels. (NAR 2X FoV is a digital zoom, it doesn chage the number of pixels actually illuminated by the object).
 
Anyway, the apparent size of Venus is so small that even if it could be seen in MWIR (which I doubt), in a 640 pixel FPA with 0.7º FoV it would be as big as ...2 pixels. (NAR 2X FoV is a digital zoom, it doesn chage the number of pixels actually illuminated by the object).

Venus is in a similar position in its orbit now as it was in Jan '15.

This is a cropped photo I took with a full spectrum camera and R720 IR filter a couple evenings ago.

1627612079889.png
 
Last edited:
Venus is in a similar position in its orbit now as it was in Jan '15.

This is a cropped photo I took with a full spectrum camera and R720 IR filter a couple evenings ago.
Nice pic. But that's in the near IR, limited to what a Silicon device (CCD) is sensible to (max. 1.1 um wavelength). Also, that's the reflection of the sun in Venus, which is still a lot of energy. Compared to the Mid-IR, its about 100 times more energy than you have in MWIR (3.7-5.0 um wavelength), where an InSb FPA is sensible.




Spectrum.png

(blue: Visible+near IR. Red: MWIR)
 
Nice pic. But that's in the near IR, limited to what a Silicon device (CCD) is sensible to (max. 1.1 um wavelength). Also, that's the reflection of the sun in Venus, which is still a lot of energy. Compared to the Mid-IR, its about 100 times more energy than you have in MWIR (3.7-5.0 um wavelength), where an InSb FPA is sensible.

The IR filter knocks down the brightness and improves contrast, which according to the article I linked in a previous post, helps to see Venus during the day.

The other trick is focus. One time I saw Venus before sunset naked eye, and what really helped was it was near a crescent moon, like in Amber Robot's photo above. I could only see Venus after my eyes adjusted to seeing the distant moon. Same goes for the camera. without something to focus on, Venus disappears against the bright sky. My photo above was taken after sunset when Venus was clearly visible.

No idea what can be seen with an InSb FPA chip. The FLIR pod has a much larger aperture (12" ?) than my camera lens, so collects far more light.
 
Last edited:
Searched the internet for mid/far infrared planet images and found but one:

1627734682501.png

Infrared image of Jupiter from SOFIA’s First Light flight composed of individual images at wavelengths of 5.4 (blue), 24 (green) and 37 microns (red) made by Cornell University’s FORCAST camera. Ground-based infrared observations are impossible at 5.4 and 37 microns and normally very difficult at 24 microns even from high mountain-top observatories such as Mauna Kea due to absorption by water and other molecules in Earth’s atmosphere.

Source: https://www.nasa.gov/content/sofia-studies-planets

I think this puts to bed any chance a FLIR camera can capture celestial objects or man made satellites outside earths atmosphere, especially an object on the horizon like in the GIMBAL video.
 
Last edited:
The 3-5 micron range is a transmission window in atmosphere for IR, with very low absorbance. That's one of the reasons(*) why it is used by ATFLIR.
But being able to image a planet I guess has more to do with very low radiation from the source, using large apertures and long exposures, and struggling with scattered light from other sources. Maybe with a dedicated equipment you can get it, but it's not ATFLIR purpouse and I would't expect it to do it.


(*)Another one is that the range includes the emission band of CO2, one of the by-products of combustion of a jet motor.
 
I'm a little confused reading through this thread. Was the airspeed of the Gimbal object ever estimated/calculated?

It seems the airspeed of the jet was calculated from what I can see. But I'm more interested in what speed the Gimbal object is doing
 
I'm a little confused reading through this thread. Was the airspeed of the Gimbal object ever estimated/calculated?

It seems the airspeed of the jet was calculated from what I can see. But I'm more interested in what speed the Gimbal object is doing

As are we but that can't be worked out without knowing the distance to it which this thread tried to work out from lines of bearing using realistic turning rates of the f/18. But the conclusions are that minor changes in the angle rate estimate change the calculations vastly the angle changes shown on the FLIR are not actually accurate enough.
 
As are we but that can't be worked out without knowing the distance to it which this thread tried to work out from lines of bearing using realistic turning rates of the f/18. But the conclusions are that minor changes in the angle rate estimate change the calculations vastly the angle changes shown on the FLIR are not actually accurate enough.
Yes, with the more detailed calculations it was shown that the lines of sight could vary from converging on a relatively near region, to being essentially parallel (meaning it could be 100s of miles away).
 
Yes, with the more detailed calculations it was shown that the lines of sight could vary from converging on a relatively near region, to being essentially parallel (meaning it could be 100s of miles away).

So the claim that the Gimbal object is stationary in the air, could be true ?
 
I'm a little confused reading through this thread. Was the airspeed of the Gimbal object ever estimated/calculated?

It seems the airspeed of the jet was calculated from what I can see. But I'm more interested in what speed the Gimbal object is doing

That's something that we have overlooked in my opinion. I have some new results I'd like to share. Here is my line of reasoning.

1) The movement of Gimbal, relative to the background (clouds), that we see, is mostly due to Gimbal motion. Parallax is secondary here because the fighter is behind Gimbal, and Gimbal goes away from the fighter. This is a very different situation than GoFast, for which the parallax effect is very important as GoFast is on the side, coming towards the fighter.

2) Because the motion we see is primarily from Gimbal, we can, very roughly, but meaningfully I think, estimate its speed from the time it takes to cross the field of view (FOV). We know it's 0.7 deg. Looking at the cloud features, a rough estimate is that it takes 2 seconds for a cloud "peak" to cross the FOV. Therefore, let's say Gimbal crosses the FOV in 2 seconds. It's not crossing it perpendicularly, but sideway, because it's seen from the back/right. From previous reconstructions, geometrical and also from flight simulations, that angle is at minimum 45deg. Even if it's less, this is only a factor 1.4 (sqrt 2) in the following calculations.

3) In GeoGebra, it's easy to create a cone with a radius that corresponds to the FOV, in function of the distance. For a given distance, the diameter of the cone, divided by the time Gimbal takes to cross it (~2sec), multipled by sqrt(2) (to account for the angle of crossing) gives us a rough speed estimate.

I've made the model here : https://www.geogebra.org/3d/xkxcpncc

The position of Gimbal can be moved, the corresponding distance to the fighter is indicated in Nautical Miles (NM).
The corresponding minimum speed (i.e. perpendicular trajectory) for Gimbal is given.

You'll see that at greater distances, the speed quickly become unrealistic for a plane. At 90 NM (~100 miles), the "distant plane" would have to go at ~5000 km/h.

"Sane" speeds (500-1000 km/h) are only found between 5 and 15Nm, which is consistent with previous estimates for the distance. A larger FOV makes for even greater speeds. I think the angle of crossing of 45deg is an underestimation, is is probably larger than that (see for example the DCS simulation of MclachlanM in the previous page). So if something the speed is likely underestimated.

Those are rough estimates, but even considering uncertainties along the way, the number I get at large distances are way beyond what is possible for a plane. I see people discussing a rocket in the other thread, would that be a plausible candidate? What is the speed of a rocket ?

Again, if it's a plane, how come its features cannot be seen for such a relatively close distance from the fighter ?

I'll be happy to correct some mistakes I may have made, I'm simply trying to fuel the discussion in a new direction and see if we can learn from it.
 
Last edited:
That's something that we have overlooked in my opinion. I have some new results I'd like to share. Here is my line of reasoning.

1) The movement of Gimbal, relative to the background (clouds), that we see, is mostly due to Gimbal motion. Parallax is secondary here because the fighter is behind Gimbal, and Gimbal goes away from the fighter. This is a very different situation than GoFast, for which the parallax effect is very important as GoFast is on the side, coming towards the fighter.

2) Because the motion we see is primarily from Gimbal, we can, very roughly, but meaningfully I think, estimate its speed from the time it takes to cross the field of view (FOV). We know it's 0.7 deg. Looking at the cloud features, a rough estimate is that it takes 2 seconds for a cloud "peak" to cross the FOV. Therefore, let's say Gimbal crosses the FOV in 2 seconds. It's not crossing it perpendicularly, but sideway, because it's seen from the back/right. From previous reconstructions, geometrical and also from flight simulations, that angle is at minimum 45deg. Even if it's less, this is a factor 1.7 (sqrt 2) in the following calculations.

3) In GeoGebra, it's easy to create a cone with a radius that corresponds to the FOV, in function of the distance. For a given distance, the diameter of the cone, divided by the time Gimbal takes to cross it (~2sec), multipled by sqrt(2) (to account for the angle of crossing) gives us a rough speed estimate.

I've made the model here : https://www.geogebra.org/3d/xkxcpncc

The position of Gimbal can be moved, the corresponding distance to the fighter is indicated in Nautical Miles (NM).
The corresponding minimum speed (i.e. perpendicular trajectory) for Gimbal is given.

You'll see that at greater distances, the speed quickly become unrealistic for a plane. At 90 NM (~100 miles), the "distant plane" would have to go at ~5000 km/h.

"Sane" speeds (500-1000 km/h) are only found between 5 and 15Nm, which is consistent with previous estimates for the distance. A larger FOV makes for even greater speeds. I think the angle of crossing of 45deg is an underestimation, is is probably larger than that (see for example the DCS simulation of MclachlanM in the previous page).

Those are rough estimates, but even considering uncertainties along the way, the number I get at large distances are way beyond what is possible for a plane. I see people discussing a rocket in the other thread, would that be a plausible candidate? What is the speed of a rocket ?

Again, if it's a plane, how come its features cannot be seen for such a relatively close distance from the fighter ?

I'll be happy to correct some mistakes I may have made, I'm simply trying to fuel the discussion in a new direction and see if we can learn from it.
There have been some developments lately about the geometry of the Gimbal video, you could check Could The Gimbal Video Show an Atlas V Launch?, starting from post #106 if you're in a hurry.
 
That's something that we have overlooked in my opinion. I have some new results I'd like to share. Here is my line of reasoning.

1) The movement of Gimbal, relative to the background (clouds), that we see, is mostly due to Gimbal motion. Parallax is secondary here because the fighter is behind Gimbal, and Gimbal goes away from the fighter. This is a very different situation than GoFast, for which the parallax effect is very important as GoFast is on the side, coming towards the fighter.

2) Because the motion we see is primarily from Gimbal, we can, very roughly, but meaningfully I think, estimate its speed from the time it takes to cross the field of view (FOV). We know it's 0.7 deg. Looking at the cloud features, a rough estimate is that it takes 2 seconds for a cloud "peak" to cross the FOV. Therefore, let's say Gimbal crosses the FOV in 2 seconds. It's not crossing it perpendicularly, but sideway, because it's seen from the back/right. From previous reconstructions, geometrical and also from flight simulations, that angle is at minimum 45deg. Even if it's less, this is only a factor 1.7 (sqrt 2) in the following calculations.

3) In GeoGebra, it's easy to create a cone with a radius that corresponds to the FOV, in function of the distance. For a given distance, the diameter of the cone, divided by the time Gimbal takes to cross it (~2sec), multipled by sqrt(2) (to account for the angle of crossing) gives us a rough speed estimate.

I've made the model here : https://www.geogebra.org/3d/xkxcpncc

The position of Gimbal can be moved, the corresponding distance to the fighter is indicated in Nautical Miles (NM).
The corresponding minimum speed (i.e. perpendicular trajectory) for Gimbal is given.

You'll see that at greater distances, the speed quickly become unrealistic for a plane. At 90 NM (~100 miles), the "distant plane" would have to go at ~5000 km/h.

"Sane" speeds (500-1000 km/h) are only found between 5 and 15Nm, which is consistent with previous estimates for the distance. A larger FOV makes for even greater speeds. I think the angle of crossing of 45deg is an underestimation, is is probably larger than that (see for example the DCS simulation of MclachlanM in the previous page). So if something the speed is likely underestimated.

Those are rough estimates, but even considering uncertainties along the way, the number I get at large distances are way beyond what is possible for a plane. I see people discussing a rocket in the other thread, would that be a plausible candidate? What is the speed of a rocket ?

Again, if it's a plane, how come its features cannot be seen for such a relatively close distance from the fighter ?

I'll be happy to correct some mistakes I may have made, I'm simply trying to fuel the discussion in a new direction and see if we can learn from it.
We already have a thread for this, the theory is that the features of a plane cannot be seen because the IR glare of the heat source is larger than the physical size of the object and obscures it or it's so far away that all we can see is the glare from the heat source.
 
@jarlrmai : That's exactly the point of my post, it cannot be very far because its speed would be unphysical for a plane.

To me there is no question anymore, it's either a relatively close object, or a supersonic object very far, although I have a hard time following the arguments for the rocket from the other discussion.
 
@jarlrmai : That's exactly the point of my post, it cannot be very far because its speed would be unphysical for a plane.

To me there is no question anymore, it's either a relatively close object, or a supersonic object very far, although I have a hard time following the arguments for the rocket from the other discussion.
Hint: it's a supersonic object very far (call it rocket if you wish).
 
Note that I used a FOV of 0.7, and because the ATFLIR is in Mode 2, it may be 0.35 deg (is it for sure?). Regardless of this, the speed at long distances are too fast for a plane.
 
Note that I used a FOV of 0.7, and because the ATFLIR is in Mode 2, it may be 0.35 deg (is it for sure?). Regardless of this, the speed at long distances are too fast for a plane.
I assume these steps:
NAR=1,5°
NAR zoom1=0,7°
NAR zoom2=0,35°
Do you think Is It correct my consideration?
 
the 2.0x is zoomed once if that makes sense

There isn't NAR then NAR Zoom level 1 and then NAR Zoom level 2

There is NAR and then NAR 2x
 
Have you considered that the clouds may be moving?

I have not. If the clouds have a speed due to difference in wind relative to the altitude of Gimbal and the fighter, I think this is secondary compared to the speed of Gimbal, and/or apparent speed due to the F18 speed (parallax effect, important at close distance, less at far distance).

While I'm at it, I want to copy what I posted in another thread because it belong more here.

This is relevant to what @MclachlanM proposed in the previous page, as a potential steady trajectory with parallel line of sights. I think it's not possible for the following reasons :

If we look at the video, the clouds move from left to right, i.e. they are scanned by the camera from right to left (or in another words we see more and more of the clouds to the left of the FOV).

Let's put this information into the equation, and consider parallel line of sights with a simple schematic. That gives us something like this :

Parallel (or at least non-crossing) line of sights
Parallel line of sights.jpg

See the problem ? If we are in this configuration, the camera should scan the clouds from left to right, i.e. we should see more and more clouds to the right of the FOV. This doesn't work.

Now, if the lines of sighting cross, it's possible to reconcile what we see with the clouds.

Crossing line of sights
Crossing line of sights.jpg

In that configuration we see more and more clouds to the left of the FOV. This is the configuration we see in the video. With crossing lines of sight, it becomes very difficult to build a steady trajectory for a distant plane that would not deviate too much from the lines of sight we have.

I think if @MclachlanM had background clouds in his simulation, we would see the clouds moving in the opposite direction to the video.
 
Last edited:
How valid is the assumption that those are background clouds and not foreground clouds?
This is a really good question. In my opinion one of the most probable clues that the clouds are in front of the object is that they are rather detailed, we can distinguish the various cumulus clouds and above all the various transversal bands typical of stratiform formations. If they had been further away it would have been less distinct.
 
If the lines of sight were parallel, clouds in the forefront would still move from left to right, wouldn't they? It doesn't change my point that the lines of sight have to cross.
 
1s to 10s is 1.4737°
10s to 20s is 1.1454°
20s to 30s is 0.6653°

Based on a 0.35° Field of View

We were on an interesting path before being distracted by the object possibly being Venus. I suggest we keep going based on what we've learnt so far.

The LoS angles that CassiO retrieved using frames (post #179 and before) are very interesting because they can help refine the geometrical reconstruction based on the ATFLIR angles and F-18 trajectory. I've included them in my model, to refine the rate of turns so they match the "frame analysis" angles. This gives something like this :

Screenshot 2021-11-23 140921.jpg

Unless somebody comes up with good arguments against it, I consider this is a good estimate of the situation. This matches the retrieval of the lines of sight using two different methods, and it is consistent with the movement of the background clouds (opening to the left).

There may be an influence of the movement of the clouds because they are not at the same altitude of the F-18, and there is probably some wind shear so they are not in the same frame of reference. But even considering a very high wind shear (120 Knots), the clouds would only cover the distance marked by the reference blue segment in the figure above. I consider this of secondary importance compared to the speed of the F-18, and the speed of a potential plane (plus the wind shear is very likely less than that).

If Gimbal was only moving against the wind (120 Knots to the west) and thus being stationary in the F-18 frame of reference, it should be in the 20-30 Nm range (23Nm in the position marked above). This very well aligns with @Edward Current estimate that the object gets 9% closer over the course of the video (see post #190), because the distance at mark 0'01 is 23.76 Nm, at mark 0'31 it is 21.2 Nm. That's ~ 10% closer.

This is of course inconsistent with a plane. For a plane to be the answer here, it has to go from one line of sight to the other behind the interception point, and cover a reasonable distance given the speed of a plane at this altitude (see reference segment for a plane flying at 500 Knots). This is not impossible, but the plane has to turn because it has to cover a same distance between each line of sight (for it to be a steady trajectory). That is what makes me doubt about the distant plane hypothesis, as we should see some kind of change in the IR glare if the engine's angle of sight changes. And it would be a coincidence that the plane turns right when the F-18 tracks it, far away from any airport. Finaly, a far away plane does not match the 9% growth in the size of the object, aka post #190 :
https://www.metabunk.org/threads/gi...lines-of-bearing-and-or-dcs.11836/post-253617

To me this starts to be too much evidence against the distant, or not so distant in fact, plane.

I've been posting a lot lately as I had time to go back to this, I will stop now and keep others share their thoughts. Thanks for reading.

EDIT : just to add that if Gimbal is between 20-30Nm from the F-18, it's 11-17m long (based on the 0.35° FOV)

EDIT 2 : the latest version of the model I use https://www.geogebra.org/m/p4zhvaaf
 
Last edited:
Unless somebody comes up with good arguments against it,
The clouds are still moving while the UAP is bearing 0⁰, they slow down but don't quite stop until the end when the UAP is bearing right. And the F-18 is still turning left at the time.
With a stationary UAP, the clouds should stop shifting at 0⁰ and then reverse direction.

Please explain how the end of the video works.

Source: https://m.youtube.com/watch?v=QKHg-vnTFsM
 
Excellent insight! While the LOS remains fixed, the object and the clouds being aligned, the azimuth continues to increase to the right.
Do you have your own hypothesis?
 
The clouds are still moving while the UAP is bearing 0⁰, they slow down but don't quite stop until the end when the UAP is bearing right. And the F-18 is still turning left at the time.
With a stationary UAP, the clouds should stop shifting at 0⁰ and then reverse direction.
Yes I noticed that too, I have two hypotheses to explain it :

- the object is not stationary, it's moving but then stops at the end of the video. I think when Ryan Graves mentions the object being stationary, he may mean that the object was flying at a low speed impossible to achieve for a regular aircraft without stalling. Or maybe he only talks about the very end of the video. So the object could be a bit before or after the point of intersection, and have a relatively slow motion during most of the video that contributes to its speed relative to the clouds (i.e. the motion of Gimbal is not only an apparent motion due to parallax). The parallax effect must stop at bearing 0⁰, but it doesn't, because Gimbal is moving. But then when Gimbal stops, the angle between the F-18 and Gimbal lines of bearing is too low to induce any significant parallax.

- Gimbal is stationary, or close, but this is an effect of clouds motion at that point, when their displacement is not secondary anymore compared to the apparent speed due to parallax.

Or maybe a bit of both.
I'll be happy to hear other's takes on this.
 
Graves, from 20:25 in the video below:
This object [gimbal] was proceeding behind the larger group and essentially it approached and then it stopped for a bit and then as the formation turned back 180 degrees it just stopped from that position and then immediately went in the other direction more or less.
 
My take... The plane is still banked left, but it's not turning anymore. That is to say, the turn was coordinated until the target was more or less front of the plane.
 
Back
Top