Jump to content

JHMCS roll alignment


SeaBass80

Recommended Posts

I am wondering about the roll-axis alignment of the JHMCS.

Observed behavior (DCS): in roll alignment mode, one is supposed to rotate clockwise or counterclockwise the orientation of the two alignment crosses to make them coincide with the vertical reference line. But if the head is tilted/rolled during this procedure, the two crosses actually again form a non-zero angle with the vertical reference line, so you don't actually know whether you are correctly roll-aligned or not.

Expected behavior (what I think it should work like - I don't have real-life experience): the relative angle between the imaginary line connecting the two alignment crosses and the vertical reference line should not change with head roll/tilt. During roll alignment, one merely has to bring this angle to zero by rotating the crosses manually clockwise or counterclockwise. When done so, the alignment crosses should stay on the vertical reference line independent of head rotation.

The way it is implemented now I guess would make it really difficult to (re)align the JHMCS during a bank turn, as your head would likely tend to align with the horizon when sitting in the cockpit for real. If one goes through the alignment procdure under these circumstances, their is a noticeable head-roll dependent pointing error in the HMCS steerpoint cue. Does anyone have any real-life experience on how these helmets are supposed to align? Unfortunately I don't have any original written references.

Link to comment
Share on other sites

This took me a minute to understand. Agreed, the point of the calibration is to refine the understanding of zero between the HUD and HMCS. The fixed registration marks are projected on the HUD while the moveable alignment marks are projected on the HMCS. The alignment marks must be projected on what the system thinks is the HUD's frame of reference. The roll alignment mark must be "head roll stabilized" or there's no point in the calibration process.

A track to demonstrate what such a looks-right-but-is-wrong alignment would look like

F-16 HMCS roll alignment.trk


Edited by Frederf
Link to comment
Share on other sites

Not a lot of people seem to mind this issue, given the lack of responses. Since the F-16 does not come pre-aligned when cold-started, proper implementation of the alignment is rather important, I think. @Frederf, so I get that you consider the missing "head roll stabilization" of the roll alignment marks as a bug as well, right? Can topics be shifted to the bug thread, or should I repost it there?

Link to comment
Share on other sites

It's just a restatement of what you said, nothing new.

Bytes are cheap. I would save the moderator effort and make a new post with as few words as possible to describe the issue. For example "movable roll alignment mark moves with pilot head tilt". Many more words than that risks confusion.

Link to comment
Share on other sites

I just saw your track file and then tried HMCS alignment on ground, then in straight and level flight and later during turn. It works fine for me, the way I believe it's supposed to do. Try doing the alignment by actually putting in a turn instead of tilting your head too much while the aircraft is still straight. You are putting a error by yourself by that head tilt.
 


Edited by Gattling
Link to comment
Share on other sites

4 hours ago, Frederf said:

The point is if the pilot tilts his head the symbology displayed shouldn't change at all relative to the cockpit frame of reference. The fact it does is a bug in itself.

The symbology should say static compared to the helmet. All the roll calibration does is account for the difference between the fitment of the helmet and the pilot holding his head upright. When the pilot tilts his head the symbology in the helmet should not move compared to the pilots head. 

Proud owner of:

PointCTRL VR : Finger Trackers for VR -- Real Simulator : FSSB R3L Force Sensing Stick. -- Deltasim : Force Sensor WH Slew Upgrade -- Mach3Ti Ring : Real Flown Mach 3 SR-71 Titanium, made into an amazing ring.

 

My Fathers Aviation Memoirs: 50 Years of Flying Fun - From Hunter to Spitfire and back again.

Link to comment
Share on other sites

No, the symbology is attempting to draw a vertically deposed set of symbols in the cockpit frame of reference. The calibration is to make agree the displayed vertical and the actual vertical. As the helmet tilts 1-2-3-4-5-etc. degrees the symbology in the helmet frame of reference tilts 1-2-3-4-5-etc. degrees the opposite way. If the roll calibration is of X degrees then it will be displayed that constant error at any helmet tilt angle. The relationship will be 1:1 with a delta angle. Calibration's purpose is to make that delta angle zero.

Link to comment
Share on other sites

@FrederfLet me understand your statement. If pilot rolls his head 45 degrees to the right while looking forward, just to the side of the HUD, you believe the symbology in the helmet should maintain level in regards to the cockpit? That is to say roll 45 degrees to the left? And by “symbology” I mean heading tape, airspeed altitude etc etc.

My brother flew the RDAF MLU Vipers and in that aircraft the HMD symbology was fixed to the helmet. That is to say it’s correct as it is in DCS.

This can be seen well at 33:00 in this video: 

 


Edited by Deano87

Proud owner of:

PointCTRL VR : Finger Trackers for VR -- Real Simulator : FSSB R3L Force Sensing Stick. -- Deltasim : Force Sensor WH Slew Upgrade -- Mach3Ti Ring : Real Flown Mach 3 SR-71 Titanium, made into an amazing ring.

 

My Fathers Aviation Memoirs: 50 Years of Flying Fun - From Hunter to Spitfire and back again.

Link to comment
Share on other sites

Ahhhh ok my mistake. I didn’t catch that’s what you meant.

I’m not sure you’re correct though, the pilot holds his head level, then you rotate the calibration crosses to line up, and then essentially calibrates how much offset the helmet is from true vertical when on the pilots head. What would be the point of the rotation step if it *Didn’t* rotate with the helmet? I’ll ask my brother, but that may be beyond his memory. 

Proud owner of:

PointCTRL VR : Finger Trackers for VR -- Real Simulator : FSSB R3L Force Sensing Stick. -- Deltasim : Force Sensor WH Slew Upgrade -- Mach3Ti Ring : Real Flown Mach 3 SR-71 Titanium, made into an amazing ring.

 

My Fathers Aviation Memoirs: 50 Years of Flying Fun - From Hunter to Spitfire and back again.

Link to comment
Share on other sites

Yes, obviously. But think about it. If the rotation calibration crosses don’t move with the helmet motion then how do you calibrate them? The whole point is that you’re calibrating the helmet when the pilot feels like he’s holding his head upright, looking forward. Due to changes in shape of the pilots head and general non symmetry of humans it’s entirely possible that the helmet may be a few degrees off from vertical. The crosses have to move with the helmet, so that the pilot can drive them back to vertical as part of the calibration process and thus establish the required offset to level the symbology inside the helmet. 

  • Like 1

Proud owner of:

PointCTRL VR : Finger Trackers for VR -- Real Simulator : FSSB R3L Force Sensing Stick. -- Deltasim : Force Sensor WH Slew Upgrade -- Mach3Ti Ring : Real Flown Mach 3 SR-71 Titanium, made into an amazing ring.

 

My Fathers Aviation Memoirs: 50 Years of Flying Fun - From Hunter to Spitfire and back again.

Link to comment
Share on other sites

To me, the main point is that the roll calibration should work with whatever head roll the pilot defines as his "center" position. In that regard it does not matter whether the pilot's neck is not straight or the helmet itself is slightly off. At the moment, we get an error that is proportional to how much head roll was used when doing the alignment. I *think" in the final step of the calibration (roll alignment), the helmet would need to project the two crosses in a way that the upper one always stays on the center of the HUD cross and the lower one always stays on the vertical line. This should be true for whatever head position the pilot uses at that moment. If it is off, it is to be corrected with the cursor slew axis until the crosses always fall on top of the vertical line.

Yes, the symboly *after" the alignment is projected with respect to the helmet "up", otherwise this would make reading stuff really difficult.

At the moment, the roll error after alignment is more or less proportional to how much head "tilt" is used during alignment. The purpose of the alignment should not be to teach the pilot how he should hold his head, but to adapt to him and the helmet. In real life, there is now tilt axis on my head that I can center, after all.

Maybe somebody knows a former pilot with hands-on experience with these helmets.

  • Like 1
Link to comment
Share on other sites

6 hours ago, Deano87 said:

Yes, obviously. But think about it. If the rotation calibration crosses don’t move with the helmet motion then how do you calibrate them?

By slewing with the radar cursor.

6 hours ago, Deano87 said:

The whole point is that you’re calibrating the helmet when the pilot feels like he’s holding his head upright, looking forward.

The point of calibration is that the helmet symbology is projected accurate to the outside world. The pilot's head "zero" is irrelevant. This isn't like TrackIR where centering is an event that assumes a particular orientation. We're smoothly adjusting a control until two images agree.

The dx dy calibration already embodies this concept. As you're adjusting agreement it doesn't matter if you yaw or pitch your head wanders. The alignment marks appear fixed relative to the HUD marks by compensating for head motion.

  • Like 1
Link to comment
Share on other sites

7 minutes ago, SeaBass80 said:

Does anybody here fly the F-18? Are the roll alignment marks in its JHMCS alignment procedure also not roll-stabilized?

I believe I misunderstood you at the first instance. Have gone through the discussion and Track file once again for more deliberate understanding. The HMCS display must stabilize in the HUD FOV during Az/El and Roll alignment modes as soon as they are mode-selected. If they are behaving in any other way (as shown in your trackfile), it's likely because of some error. 


Edited by Gattling
Link to comment
Share on other sites

  • ED Team
2 hours ago, Frederf said:

By slewing with the radar cursor.

The point of calibration is that the helmet symbology is projected accurate to the outside world. The pilot's head "zero" is irrelevant. This isn't like TrackIR where centering is an event that assumes a particular orientation. We're smoothly adjusting a control until two images agree.

Could you provide any reference to prove your point of view, please?

Link to comment
Share on other sites

I guess we were hoping for some SME to magically chip in and provide hands-on experience 😉

As far as I know, the JHMCS uses a magnetic field (three orthogonal coils with offset AC currents) projected by the MTU unit mounted to the canopy to track its position. There really is no "zero" position for any axis unless properly calibrated with closed canopy. JHMCS roll axis calibration should work regardless of the actual in-game roll-axis position. At the moment, it only produces correct roll calibration if the in-game roll axis is at zero during calibration. To bring home that point: for the JHMCS calibration in the F-16, it doesn't matter where in XYZ space you sit during the DX/DY calibration, and it does not matter whether you stare exactly straight into the HUD or tilt your head a little during DX/DY calibration. For roll calibration though, you must have zero roll on the joystick/trackir axis to a get good calibration. To non-SMEs, this just seems wrong.

Link to comment
Share on other sites

Maybe the only written reference I found is here, in the F-16A/B Mid-Life Update tape 3 documentation , page 63. Obviously not the same aircraft model, though my guess is that the JHMCS is similar. The DED menu certainly has the same HMCS calibration procedure. The manual says that upon entering either the DX/DY or the roll alignment procedures, "The HMCS display stabilizes on the HUD". At the moment, this only works for the XYZ and azimuth/elevation head coordinates, but not for head roll. Maybe this helps as a reference for the development team to take a look.

Link to comment
Share on other sites

That would be the operative sentence in the -34 too. Those "change log" type documents are written sourced from the -34s as evidenced by the same six lines in the same order. That's as good as one is going to get out of an operating manual. For a persuasive inductive (not deductive) argument I point to the az/el mode which shares the same "is stabilized" description. Only adjustment by slewing the cursor control changes the relative position of the helmet/HUD symbols. Yaw and pitch of the helmet doesn't. Roll is the same.

Link to comment
Share on other sites

  • ED Team
On 4/18/2022 at 3:22 AM, Frederf said:

No, the symbology is attempting to draw a vertically deposed set of symbols in the cockpit frame of reference...

I may have misunderstood your sentence. I thought that you suggest that the whole HMCS indication will rotate to be roll stabilized. But after a few readings, I assume that you were only talking about the indication of calibration? If so, it sounds reasonable, but we have to make research to be sure.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

@cofcorpse Yes, we are only talking about the calibration marks. During the DX/DY and roll calibration steps, the JHMCS helmet should project calibration marks into the field of view that are "stabilized on the HUD", i.e. they will not move relative to the HUD symbols if the JHMCS is moved (X,Y,Z) or rotated (yaw, pitch, roll) in any way, as long as the HUD stays in view. The calibration consists in moving these symbols into the correct DX/DY position and the correct roll orientation by the slew cursor only. Head movement should not move them at all during the calibration step.

The fact the HMCS calibration marks move with head roll likely indicates a problem in how the roll and roll calibration is implemented in the DCS JHMCS - at the moment, it cannot produce a calibrated helmet if roll axis is not zero, which contradicts the purpose of the calibration.

  • Like 4
Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...