Night Vision Clip-ELR collimation

[CoryT]

Testing the ELR, it seems to be very sensitive to position. Placing pressure on the fore end shifts POI. if I turn the reticle on and center it on the crosshairs, any pressure on the rail or displacement of the unit moves the crosshairs off the internal reticle. On and off the rail, or forward and back a slot also changes the reticle position. I’ve tried all three modes, same thing. 3 different rifles, same thing. Ran through full range of parallax still movement.

It‘s not small, more than .8 mil of movement. Has anyone else tested or seen this?

 

[CoryT]

Further investigation reveals the menus and cross projection are NOT in the same plane as the image, and do not collimate correctly. The image seems far more stable. I will test on the collimation table tomorrow and see what I can measure for boresight error. Test group this evening at 300 was on for elevation but .3 mil right.

 

[CoryT]

Bore sight error measures .2 mil, which is perfectly reasonable and within spec. It is less tolerant of axial misalignment than any of my I2 clip on devices, but there is some allowance for planar misalignment, up/down or left right by 1/2 inch shows no additional shift. 5 degrees off axis is as much as 1 mil of error, so,I would advise a 1 piece rail and a flat mount, or be very sure the rails are in good alignment.

 

[Conqueror]

Cory, what have you seen on other high-end units? From an optical standpoint it doesn’t seem crazy to me that any clip-on would suffer from image shift if it is axially misaligned (ie twisted 5 degrees on a bad rail). But I have only ever run clip-ons on monolithic rails or quality short AR rails that don’t have much flex.

 

[CoryT]

Most of the I2 type units display very little boresight error even when handheld in front of the day optic. Those are normally using a split (Risley) prism assembly, and once set, if you can see through it you are pretty much good to go. For whatever reason, thermals don't seem to be done the same way, either having no boresight in the first place and just zeroing the screen to the rifle, or having a lens system with less error margin.

 

[Conqueror]

Yeah my current understanding is that thermals don’t use Risley prisms, but that their collimation is done digitally in some kind of “pixel mapping” operation at the factory. I feel there must be some engineering reason for this beyond cost-savings since this is even true on no-budget systems intended for military sales. I’m no optician but I imagine it must be challenging to invent a rear eyepiece that shows a digital screen identically despite looking through it from different positions and angles. To me, repeatability is paramount over absolute shift, though I’d also like absolute shift as small as possible.

 

[CoryT]

I’m going to pose that question to some L3 engineers that will be here for class next week. Clearly there is some reason for it, which most probably relates to the differences in the two display types, phosphor screen vs digital OLED or similar display, which is really quite a bit lower resolution.

 

[cantankerous]

I think there are a number of us who would love to hear how the "collimation" works!l

 

[rlsmith1]

I'd guess the Riley prism has some sort of patent on it and may not be available for use... WOuld love to hear more on collimation as well.

So what I'm hearing you say is a 20 MOA receiver will have some offset if used with a 0 MOA handguard? The solution to this would be to purchase a 20 MOA mount and use the mount backwards to cancel out the elevation.

ETA: glass adds weight and length too

 

[cantankerous]

I'd guess the Riley prism has some sort of patent on it and may not be available for use... WOuld love to hear more on collimation as well.

So what I'm hearing you say is a 20 MOA receiver will have some offset if used with a 0 MOA handguard? The solution to this would be to purchase a 20 MOA mount and use the mount backwards to cancel out the elevation.

ETA: glass adds weight and length too

This makes sense and explains the 2 MOA shift down I get with my ELR on my RPR. 20 MOA scope rail and 0 MOA free float tube.
It shoots POA/POI on my AR.

 

[CoryT]

The Risley prism system it not patent controlled. It’s commonly used in beam steering applications, of which collimation of a night vision sight is a good usage, as they are compact and not sensitive to vibration.

The angled base with flat rail is one of the angular alignment problems that will cause a shift in these thermals where a prism alignment would fix it.

 

[rlsmith1]

Curious why more don’t use the prism then…

 

[Ksracer]

I wonder how the stepped design of these effects things?

When I lock my rifle down and introduce the thermal, it raises the crosshairs about an inch on the target, which is roughly equal to to amount of the offset between the front and rear lenses. My grouping with the thermal on is about an inch low, which would reflect that. My theory is that it is a PARALLEL offset from the line of sight and remains 1" from 100yds to infinity, based on the limited shooting I've done. I also believe that if you were to re-zero your rifle with the thermal on, you would then need to adjust the bore-height in your ballistic calculator to match up with the thermals front lens. Just my wild ass theory.

For what it's worth, I can leave the thermals clamp loose and wobble it to the extremes of travel and the image doesn't move appreciably under the crosshairs.

Regardless, I can deal with any minor boresight errors as long as they are consistent.

 

[Conqueror]

In the era of the Kestrel and the Raptar the need for absolute minimal shift may be fading? I can just save a rifle profile with a zero offset of whatever the thermal POI shift is and get right back to shooting as long as it’s repeatable. And for shorter range shooting it’s not that hard to just dial a 0.3 mil change when the thermal goes on and then just use reticle holds. Consistency is key.

 

[cantankerous]

I wonder how the stepped design of these effects things?

When I lock my rifle down and introduce the thermal, it raises the crosshairs about an inch on the target, which is roughly equal to to amount of the offset between the front and rear lenses. My grouping with the thermal on is about an inch low, which would reflect that. My theory is that it is a PARALLEL offset from the line of sight and remains 1" from 100yds to infinity, based on the limited shooting I've done. I also believe that if you were to re-zero your rifle with the thermal on, you would then need to adjust the bore-height in your ballistic calculator to match up with the thermals front lens. Just my wild ass theory.

For what it's worth, I can leave the thermals clamp loose and wobble it to the extremes of travel and the image doesn't move appreciably under the crosshairs.

Regardless, I can deal with any minor boresight errors as long as they are consistent.

What you described is what happens when you use a SIMRAD.
It sits like 3.5 inches higher than your day scope.

your POI will be 3.5 inches lower (linear not angular) than with just your day scope at every distance.

 

[Ksracer]

What you described is what happens when you use a SIMRAD.
It sits like 3.5 inches higher than your day scope.

your POI will be 3.5 inches lower (linear not angular) than with just your day scope at every distance.

What adjustments, if any, do you make to compensate for it?

 

[Conqueror]

It sounds like you shouldn’t make any. Just dial your normal dope and hold 3” high on every target regardless of distance (or just hold center since that’s a meaningless difference on a man-size target within 1000 yards).

 

[cantankerous]

I don't think the military did adjust for it.

Personally I would adjust the extra 3.5 inches up @ 100yrds and it gave the appearance of an extremely flat shooting cartridge. Yes it changed my dope but I wasn't shooting it much passed 450 anyway.

 

[Ksracer]

Was able to spend some more time at the range this morning. First thing I did was check my 100yd zero and chronograph. I adjusted my Kestrel by 30fps. That fixed my previous dope issue at 700yds.

Then I shot back to back groups at 100, 400, and 700 yds each.

I still need to tweak my 400yd dope, but I'm pretty happy with the minimal shift I'm seeing with the thermal.

My 100yd group clipped the lower left corner of the hothands. So almost exactly 1" low from my zero. Nearly identical to the previous time.

The shots at 400yds landed so close it was impossible for my spotter to tell which ones were which.

These are 64gn TGK's running 3720fps from a 22creed. Not really ideal for a 700yd shot, but most of these impacts would have resulted in a dead coyote.