PDA

View Full Version : Proper Use of Aperture Sights


ks_shooter
May 21, 2007, 06:29 PM
I remember the first time I used the aperture sights on my AR-15, I felt that they were the worst sights ever invented. Compared to the sights I was used to, there seemed no way to accurately align the front and rear sight. Books and internet sites confidently asserted that "the eye will naturally center the post" in the rear sight, so I forged ahead.

I eventually learned to shoot my AR accurately (well enough to earn my Highpower Master classification), but I never fully accepted this piece of wisdom about aiming with aperture sights. Over the last several years I have made a study of aperture sights and eventually came to some startling conclusions. Sight alignment, as normally taught, is of little importance when aiming with aperture sights - as long as the aperture is relatively small. By small I mean less than 0.100", or so. Larger apertures (ghost rings) require more attention to sight alignment, but still are amazingly forgiving of sight misalignment. It turns out that aperture sights of the target variety suppress parallax that normally causes aiming errors when using open sights.

A friend and I have detailed the reasons for this phenomenon in a paper that we have posted on the Web:

http://doug.kerr.home.att.net/pumpkin/#ApertureSight

Dave Haven
May 21, 2007, 10:10 PM
I don't look at the rear sight, I look through it and align the front sight with the target. Your eye will automatically find the center of the aperture.

Dave R
May 21, 2007, 10:47 PM
Wow, your article is a great explanation of why aperture sights work. I only know that they do work. And its nice not to have to align 3 planes--rear sight, front sight, and target. Apertures only require aligning the front post and target.

ks_shooter
May 21, 2007, 10:48 PM
Dave,

I'm 44, I couldn't look AT the rear sight if I wanted to. I have a hard enough time focusing on the front sight! I agree 100% with your description of looking through the rear sight and concentrating on the front sight and target.

The point I am trying to make is that it is usually taught that sight alignment is critical. It is also usually offered that the eye can more or less "naturally" align the sights to an adequate level. I can demonstrate that this understanding is flawed.

The truth is that sight alignment is not critical, and the the ability of the eye to center the front sight is good, but it is not good enough. If it is believed that sight alignment is as sensitive as sight picture, then one would need to center the front sight within the aperture to within a fraction of a width of the front sight. This would not happen "naturally".

But, since sight alignment is not critical, even casual attention to centering the front sight is adequate. Many shooters expend a lot of effort attempting to precisely center the front sight in the aperture, and are not accomplishing anything.

Correct understanding of this fact allows a shooter (particularly new ones) not to fuss over sight alignment. "More or less in the center" is good enough.
It sounds like you already use this approach. Many do not.


Robert

ks_shooter
May 23, 2007, 09:56 PM
As a teaser, I offer the following simple experiment that can be done to observe the lack of error when aiming with aperture sights that are "misaligned":

1) Place a rifle with suitable aperture sights (aperture less than 0.100") on a stable shooting rest.

2) Adjust the position of the rifle so that the front sight is aligned with the target, but do so with the front sight offset from the center of the aperture a generous amount, making sure that the front sight is still clearly visible (not touching the "fuzzy edge" of the aperture).

3) Without moving the rifle, adjust your head position so that the front sight appears centered in the aperture.

4) Observe the alignment of the target and front sight.

If the final "centered up" sight picture has the front sight and target properly aligned, then it can be concluded that the position of the front sight within the aperture does not effect the aiming process. The initial misalignment of the sights did not cause any parallax shift between the front sight and target, so the resulting "aim" was still correct.

With open sights, if you misalign the sights, align the front sight to the target, and then (with out moving the rifle) re-locate your eye so that the front and rear sights once again appear aligned, you will notice that the front sight and target are no longer aligned. The parallax from the misaligned sights caused the initial aim to be in error. With (small) aperture sights this does not occur.

Axion
May 24, 2007, 01:52 AM
Bookmarked you writeup for reading later.

MythBuster
May 24, 2007, 07:53 AM
The aperture front sight is also the best way to go for shooting paper.

My scores in smallbore went up after changing from a post front sight to an aperture.

ks_shooter
May 24, 2007, 12:49 PM
Mythbuster,

Aperture front sights are definitely superior for target shooting. No need to try and discern where the edge of a fuzzy target is in order to align it with the front sight.

The concept being discussed here applies directly to aperture front sights also. In this case, the centering of the front globe in the rear aperture is the activity in mind. It is usually indicated that there must be an equal amount of white around the globe for the sights to be properly aligned. Because of the parallax suppresion provided by the aperture, this is not actually required for proper aiming.

Sight alignment is also an indicator of head position which is important in terms of maintaining a consistent hold, so large amounts of sight offset may affect POI through the secondary effect of changing the hold on the rifle.

Globe front sights are not recommended for self-defense since the field of view is obstructed. The advantage of an aperture front sight can be maintained, without the use of a globe, by using an "aperture-post" front sight. Considering the parallax suppression effect, the globe is not required for adequate sight alignment, anyway.

I have constructed such a sight by soldering a small tube onto the top of a filed down AR-15 front sight (similar to the Levang sight sold by DPMS, but with a smaller diameter aperture). It worked very well. For small targets the diameter of the aperture was sized to provide a typical sight picture with the target centered in the aperture. For large targets the front aperture was used like a "bead" front sight and merely placed over the target.

Jason_G
May 24, 2007, 01:35 PM
I've always wondered if the front aperture idea would work on handguns. I have a USP that has Novak's sights on it, and the front is a FO. I always wondered how it would do if I just pushed out the FO insert and left the hole in the front post. I know this is the rifles section, but since we're talking apertures...

Jason

ks_shooter
May 24, 2007, 02:31 PM
Jason,

With a handgun the aperture will need to be fairly large in order to be able to see through it at arms length. If the aperture is larger than a couple of millimeters, it will be too large for the beneficial effect I describe to take place. The aperture needs to be smaller than the pupil of the eye.

fisherman66
May 24, 2007, 02:42 PM
The aperture needs to be smaller than the pupil of the eye.

Isn't that why there is no parallax to correct for?

Jason_G
May 24, 2007, 02:43 PM
Hmmm, I'm guessing the hole would be about 1/16" or so.

Jason

ks_shooter
May 24, 2007, 02:52 PM
fisherman66,

That is correct.

One way to look at this phenomenon is to consider the rear sight aperture and the eye as a "unit". The sight aperture replaces the pupil in defining which light rays will enter the eye, and therefore defines the point of perspective, or "what point you seem to be looking from". This doesn't change when the eye moves (within limits) because the subsititute pupil (the sight aperture) hasn't moved.

Jimro
May 24, 2007, 02:55 PM
Parallax has to do with the plane of focus verses the plane of the target. When the aiming reticle and target are on the same focal plane then there is no parallax. Since the only focal plane you have to deal with using iron sights is the one created by your eye you don't have to worry about parallax. With a scope you have to worry about the target focal plane and the reticle plane.

As far as your eye naturally centering the front sight post in the rear arpeture, I don't think so. Learning to achieve a consistent sight picture is the harder part of BRM to teach, sight picture should equal sight alignment in a perfect world. Drill Sergeants still teach the "nose to charging handle" method, the AMU teaches "where your head is naturally comfortable" method. Both work, but I find it easier to put my head where it falls naturally on the stock.

Jimro

ks_shooter
May 24, 2007, 02:57 PM
Jason,

Sorry I thought you were talking about a rear aperture on a handgun. A front aperture might be useful by offering an improved sight picture, but it wouldn't have the benefit of being insensitive to sight misalgnment.

ks_shooter
May 24, 2007, 03:07 PM
Jimro,

That is one particular definition of parallax. The one I am using is the one that comes from Webster's:

"the apparent displacement or the difference in apparent direction of an object as seen from two different points not on a straight line with the object"

In the context of our paper, we define parallax as the apparent displacement of the sights and target from each other as they are viewed from different points.

If a perfect sight picture is obtained with normal open sights, and then the eye is moved slightly off-axis, the sights and the target no longer appear to be in line with each other (even though they actually are). This is what we are calling parallax. If the sights are left unaligned and the front sight is again aligned with the target, the aim will actually be wrong, even though it "looks" right (at least the alignment of the front sight and target). In this way parallax fools one into mis-aiming. With small apertures this parallax effect doesn't occur.

Jimro
May 24, 2007, 04:16 PM
If a perfect sight picture is obtained with normal open sights, and then the eye is moved slightly off-axis, the sights and the target no longer appear to be in line with each other (even though they actually are). This is what we are calling parallax. If the sights are left unaligned and the front sight is again aligned with the target, the aim will actually be wrong, even though it "looks" right (at least the alignment of the front sight and target). In this way parallax fools one into mis-aiming. With small apertures this parallax effect doesn't occur.


What you are calling parallax is what we refer to as sight alignment and sight picture. Small apertures don't eliminate errors in sight aligment, they minimize them. The sights set up a geometric relationship. It is a cone, and the aperture is the base of the cone, the front sight post is the tip.

Large sights have a larger base and therefor more room to make mistakes, small apertures have less room to make mistakes. When I got to shoot at the AMU's 300 meter international range I found that the sights on the match rifle I was shooting always kept me on the target, but that if I didn't concentrate on getting the exact same sight picture every time my group would wander around the 10 ring, sometimes to the top right, sometimes to the bottom left.

Anyways, too small of an aperture and you don't let enough light through to get a good sight picture, too big and it's difficult to center the front sight post properly.

Jimro

ks_shooter
May 24, 2007, 04:30 PM
Small apertures don't eliminate errors in sight aligment, they minimize them. The sights set up a geometric relationship. It is a cone, and the aperture is the base of the cone, the front sight post is the tip.

Jimro,

I believe you have missed the whole point of our study. Have you taken the time to read the paper that I linked to, or attempted the simple experiment that I described in one of my previous posts? I am not trying to be sarcastic. I believe that if you read the paper or try the experiment you will realize the effect is real.

The statement quoted above is certainly the accepted conventional wisdom. It is also wrong.

ks_shooter
May 25, 2007, 04:24 PM
Jimro,

I first discovered the effect being discussed while trying to determine the amount of POI shift caused by sight misalignment. I used the same analysis you are describing. The problem with the analysis is that it only uses "line of sight" reasoning and ignores the fact that the eye collects light over the entire surface of the pupil.

The lines drawn from the front sight to the edges of the aperture (the cone you described) are part of a bundle of light rays that reach the eye simultaneously. It isn't a matter of positioning the eye so it selects one of the lines or the other.

Until the spot of light on the pupil (that results from the cone of light passing through the rear aperture) moves beyond the edges of the pupil, the image the eye perceives remains the same. The perceived location of the image (where the front sight is within the aperture) may appear to move, but the image itself is the same. The image in this case being the front sight AND the target.

You really should try the sighting experiment. It is simple to do, and the effect is obvious.

Jimro
May 25, 2007, 09:59 PM
ks_shooter,

I've shot iron sights at 300 meters and have seen the effects of an inconsistant sight picture on the target. This is using a match rifle and match grade ammunition. Amazingly when I transitioned to the ACOG my groups stopped wandering around the target.

I'm not talking a huge amount of wandering, only around 2 to 3 minutes worth, but it was enough to knock my groups out of the ten ring.

As far as a simple experiment you might like to try, use different sizes of apertures and do the same "experiment" that Burdge executed but also at a greater distance. Precision shooting with the night aperture is....difficult.

Of course my experience doesn't sound as impressive as an article written by an engineer, even if the "experiment" described lacked a control and had a very small sample size.

Jimro

ks_shooter
May 25, 2007, 11:22 PM
Jimro,

My name is Robert Burdge. I indicated in my first post that I was one of the authors of the paper, but I guess I didn't say which one. I was the person who conducted the firing test for the paper. As I stated before, I have been competing in NRA Highpower for 5 years. I have thousands of rounds downrange at 300 to 600 yards. I even managed to shoot a 196-10x at 600 yards at Camp Perry last year. I am not a novice to rifle shooting.

It was while trying to improve my scores that I started looking into the question of sight alignment. I had always wondered if some of those 9's where the result of sight misalignment, as opposed to variances in my hold or poor trigger control.

I discovered the fact that the Point of Aim followed the front sight regardless of location within the aperture. This is 100% demonstrable by the aiming experiment that I detailed previously in this thread. If you own a rifle equipped with aperture sights you can perform this experiment in about 5 minutes and proof it to yourself.

A similar, but slightly different question, is whether Point of Impact follows the front sight in the same manner. Differences in sight alignment are evidence of a different head position. This change in head position could change the POI without changing the POA. This is nearly impossible to quantify. I don't know of a scientific way to predict the change in POI for a given change in hold.

What I do know, is that my experience shows that a small amount of sight misalignment does not show up as shift in POI on the target. I have evidence to the same effect from Service Rifle shooters that are much better than I am.

Concerning large apertures, the aperture needs to be substantially smaller than the pupil of the eye or the POA will track with sight alignment. Apertures smaller than about 0.100" will not suffer from misalignment problems.

I never intended to offer the small experiment in our paper as Proof of the effect. The proof is provided in detailed ray-trace analysis in the Appendix of the paper. The experiment was simply to offer some objective data for people to look at. I would encourage anyone who has read the paper to try the simple live-fire experiment for themselves. If enough people did so, I believe the evidence would support the truth of our claims.

I am not surprised that you groups shrank when using an optical sight. Even without error from sight misalignment, aiming with iron sights is always more difficult than aiming with a set of cross-hairs.

fisherman66
May 26, 2007, 07:02 AM
I appreciate that you are sharing your work Robert. I have only shot the peep a few times, but I have a William peep siting in my web cart at Midway. I think I understand what you are describing. It reminds me of the old pinhole cameras (or any shuttered camera, really). I will try your experiment when I take delivery of my rear site (I don't know if my old 30/30 is accurate enough to make a difference, but I suspect it is given the correct size rear peep; that is LARGE).

Can you move the gun to accomplish the same thing you are doing with your head (move the front site off center, but within aperture)? How are you keeping misalignment and parallax separate? Lining up the target centered and then moving your head doesn't alleviate the time spent getting the correct site picture, and in an unviced gun begs for muzzle drift (or any gun movement). I understand the concept of removing parallax, but I don't understand the practical application.

Thanks again for sharing your work. Perhaps this would make more sense to me if I had the pedigree of a competitive shooter.

Jimro
May 26, 2007, 08:55 AM
ks_shooter,

Next time I have the opportunity I will conduct the experiment you describe.

Jimro

ks_shooter
May 26, 2007, 09:19 PM
fisherman66,

If you align the sights initially on the target and then move the rifle instead of your head, you will get a different result. In this case you will observe the front sight position moving within the aperture, but you will also notice the front sight and target "separating". And this is how it should be since the actual Point of Aim has changed. The front sight is now marking the new point of aim, which is no longer on the target.

The practical application is in how to budget your time and energy when shooting. During rapid fire stages being able to essentially neglect the sight alignment (somewhere in the middle of aperture is good enough) saves time. During slow-fire stages it allows the eye to work less. The time that might have been taken "dressing up" the sights alignment can be spent dressing up the sight picture where it will do much more good.

Dave Haven
May 27, 2007, 10:38 PM
As a teaser, I offer the following simple experiment that can be done to observe the lack of error when aiming with aperture sights that are "misaligned":

1) Place a rifle with suitable aperture sights (aperture less than 0.100") on a stable shooting rest.

2) Adjust the position of the rifle so that the front sight is aligned with the target, but do so with the front sight offset from the center of the aperture a generous amount, making sure that the front sight is still clearly visible (not touching the "fuzzy edge" of the aperture).

3) Without moving the rifle, adjust your head position so that the front sight appears centered in the aperture.

4) Observe the alignment of the target and front sight.
I inadvertently did a reasonable approximation of that experiment about 30 years ago when I was working for my Dad's gunsmithing business.
I mounted a receiver sight on a rifle and proceded to bore-site it with a Sweany Site-A-Line optical collimator. As I adjusted the windage and elevation, the collimator reticle moved in relation to the front sight, even though I wasn't moving my eye in relation to the rear sight. Moving my eye in relation to the rear sight didn't cause the collimator reticle to move in relation to the front sight unless I got into the "fuzzy" area.

That said, I've shot 1-9/16" 5-shot groups at 100 yards with my Smith Corona '03A3 without consciously trying to center the front sight in the rear aperture.:)

ks_shooter
May 27, 2007, 10:54 PM
As I adjusted the windage and elevation, the collimator reticle moved in relation to the front sight, even though I wasn't moving my eye in relation to the rear sight. Moving my eye in relation to the rear sight didn't cause the collimator reticle to move in relation to the front sight.

Bingo. Thanks for info, Dave.

I have raised this subject on a variety of forums in addition to this one. I have received quite a few responses from people saying that they had seen this phenomenon in the past. I never really expected that I was the first to have noticed this.

What I couldn't find was anyone who had offer an explanation of what caused it. A Palma Team member told me that he remembers reading about a military range-test investigating this. He said it was written up in the Dope Bag section of an American Rifleman issue back in the '60's. I would love to get a hold of a copy of the magazine and see what it said.

fisherman66
May 28, 2007, 08:22 AM
I hope you will stick around KS. I enjoy understand why things work the way they do. I understand you have a good background in optics. I'd like to know your opinions on rifle scopes.

ks_shooter
May 28, 2007, 08:53 AM
fisherman66,

I can certainly offer my opinion if you have a particular scope model in mind.

I was fairly active on this board back 2001-2002. Then it closed up shop for a while. I migrated to the The High Road forum. Soon after that are started shooting NRA Highpower and now spend most of my time Web time at the NationalMatch.us forum.

fisherman66
May 28, 2007, 09:03 AM
I am looking for a scope for a Ruger #1 RSI. Because of the barrel mounted quarter rib and the falling block design I will need a fair amount of eye relief, but not an IER scope. The greatest eye relief I have found is on a Leupy VX_II 2x7-33mm. I don't want a large objective scope on a 36" rifle. It would look odd. Any thoughts/alternatives?

ks_shooter
May 28, 2007, 09:39 AM
fisherman66,

I can't give you any specific advice since I don't have a Ruger #1. I do have a Bushnell Elite 3200 on my deer rifle and it has about 4.3" of eye-relief. Very generous. I would recommend you post a query on this forum. There are lots of proud Ruger #1 owners that can help you out.

Jimro
May 28, 2007, 09:36 PM
ks_shooter,

I checked with an optics guy I trust, he says the ray trace is sound and that the most likely reason my experience doesn't jive with yours is that my rear aperture was milspec diameter instead of being 1.0mm (0.039 inches, pretty close to 0.040 aperture size preferred by most competitors I know). This was mentioned in the article but I didn't put two and two together until another set of eyes pointed out the obvious.

Jimro

ks_shooter
May 28, 2007, 10:43 PM
Jimro,

My Bushmaster upper has a standard A2 rear sight. The small aperture measures about 0.080" (about 2 mm). Is the Mil-Spec aperture you mentioned about this size? The larger aperture will have a smaller "parallax-free zone" within the aperture field of view than a smaller aperture

Doug Kerr and I are currently working on a way to identify the edge of this zone when viewing the sight picture. We can predict how much head shift will bring on parallax, but translating this into a position of the front sight within the aperture is proving to be tricky. Of course, it can be determined experimentally with a steady rifle rest, but we are looking for a method that could be used more easily. When we have a answer I will report back on this thread.

fisherman66
May 29, 2007, 07:15 AM
Do you expect it to vary based on how much your pupil dialates? How are you planning to measure that?

Jimro
May 29, 2007, 07:36 AM
ks_shooter, I've never measured a rack grade rear sight, but I just did a quick measurement on a stock Colt rear sight I have on an HBAR. Using 0.7mm pencil leads as a measuring device isn't very scientific but my calipers are back home with the rest of my tools :( Anyways two pencil leads fit through the smaller aperture with some wiggle room, so I would guesstimate that the rear sight is about 1.5mm

Jimro

ks_shooter
May 29, 2007, 08:34 AM
fisherman66,

My co-author, Doug Kerr, is currently doing experiments using a camera in place of the eye. That way we can control the size of the "pupil" (the camera aperture) and try different sizes. It also allows the sight picture to be captured as an image that can be viewed by others for discussion. Doug is working on the test now and we hope to publish the results soon.

ks_shooter
May 29, 2007, 08:35 AM
Jimro,

If the aperture diameter is about 1.5 mm, this should be enough smaller than the pupil (usually 2 to 8 mm depending on the ambient lighting) for the parallax suppression to be noticeable.

fisherman66
May 29, 2007, 08:35 AM
That's pretty sharp. I look forward to the results.

ks_shooter
May 30, 2007, 06:05 PM
I am re-posting the link to the paper that Doug and I wrote. The paper has been updated to include tests that use a camera to investigate the phenomenon. Doug performed all of the setup and testing, and the results conclusively demonstrate that the effect of parallax suppression is real. It also provides a convenient set of images to illustrate the phenomenon for the reader. The new information is in the section titled “Optical Model Tests”:

http://doug.kerr.home.att.net/pumpkin/#ApertureSight

A detailed explanation of the methods and procedures used in the camera testing are contained in a companion paper:

http://doug.kerr.home.att.net/pumpkin/#ApertureSightDemo

ks_shooter
June 2, 2007, 06:26 PM
While discussing this topic on another forum, a member raised the question of why sight adjustments work if eye movement doesn't cause a shift in the Point of Aim. After some thought I gave the following reply, which I think adds another piece to the puzzle:

"In thinking about your question again this morning I had an "epiphany" of sorts. It occurred to me that the fact that 1/4 minute sight adjustments actually work is strong evidence that sight alignment doesn't matter. For a 1/4 minute click to move the Point of Impact a true 1/4 minute on the target (assuming the conventional wisdom about the need for sight alignment is true) the sight alignment would have to be consistent to with a fraction of a 1/4 minute. Otherwise sight adjustments would get swamped out by variations in sight alignment.

To make a 1/4 minute sight adjustment work accurately, sight alignment would have to be maintained to a level of perhaps 1/10 of a minute or less. I don't think anyone can reliably center the front sight within 1/10 of a minute from shot to shot. The typical field of view through a rear sight is something like 100 MOA. To center the front sight to within 1/10 of minute would require a placement precision of 0.1% within in the aperture. This could also be thought of as requiring the front sight to be centered to within 1/60 of its width. I don't think this is possible by mortal man."

ks_shooter
June 11, 2007, 12:02 PM
My comments in Post #39 are incorrect. After thinking about it some more I realize that sight alignment precision would not need to be small compared to the size of the sight adjustments. Sorry for any confusion this generated.