[Laser] Re: Laser Digest, Vol 45, Issue 9

[Clint Turner]

Hello Tom, Following the KISS principle, I would heartily agree with 
Jim's suggestion of using a piece of white paper and moving it 
back-and-forth for the smallest spot. I would expand this method by 
simply pointing the Fresnel lens out across the landscape during the day 
(keeping the lens shaded from direct sunlight) and focusing the image of 
the scene onto the piece of paper: I found the focal length of some 
cheap, vinyl page magnifiers this way. This may or may not apply to you, 
but I'm not sure of the source of the Fresnel lenses that you are using, 
but its worth mentioning that some surplus lenses are unusable for this 
application - in particular, those used for overhead projectors: Many of 
these simply will not focus to a tight spot, making them unusable for 
either receiving or, due to their inability to form a parallel beam, 
transmitting. *** Last year, I did some testing using some high-quality 
Fresnel lenses to determine their ability to focus to a point. For all 
four of the lenses that I tested (molded optical-acrylic devices) I was 
able to focus a star (Vega) to a spot less than a millimeter in 
diameter: Roughly speaking, the size of the spot was 1/1500th of the 
focal length or smaller: Results of those tests may be viewed here: 
http://modulatedlight.org/optical_comms/fresnel_spot_sizes1.jpg The star 
Vega was chosen because it is, for all practical purposes, at infinity 
and a point source and thus has parallel beams. I have noted that with a 
good-quality, moderate-sized Fresnel lens (12" across) that 160 meters 
is still close enough that there is *still* a slight difference in focus 
between that distance and infinity - but it is small enough to be 
ignored in practical use. I have constructed an optical transceiver 
using a pair of vinyl Fresnel page magnifier lenses. While I have yet to 
measure their performance in terms of "spot size", I have successfully 
used this unit for LED/Laser comms at a distance of 172km and they 
compare reasonably well with high-quality molded acrylic lenses in terms 
of overall light gathering ability, but with somewhat inferior off-axis 
rejection due to increased scattering - something that is more important 
when used for reception in the presence of other light sources. *** 
Based on testing, I've found that with good-quality lenses, that a 
detector of about 1mm square is large enough capture all of the light 
from the distance source: One problem is that the most common, cheap PIN 
photodiodes (such as BPW34's) are much larger than that. The major 
downside of this is that response is slowed (higher device capacitance) 
and sensitivity may be reduced somewhat owing to additional device noise 
as well as the capacitive effect: This also ignores the fact that with a 
larger device, there is a larger field-of-view - something that could 
also affect sensitivity. (Detector masking can mitigate this last 
problem.) *** For receive-system focusing, I have used a simple circuit 
consisting of a modulated LED placed as far away as practical (over 500 
feet away in my case.) My circuit consists of a 4060 CMOS 
oscillator/counter with a 500kHz ceramic resonator with the LED 
connected via a potentiometer (for LED drive adjustment) to the 
divide-by-512 output. One simply receives the 976 Hz tone at the 
receiver and adjusts the focus for maximum signal: I used a laptop and 
the Spectran for this. All of this was done in an area devoid of urban 
lighting. *** Answering your question: Yes, the entire Fresnel should 
really be illuminated for best results when focusing. You also touch on 
another point when you mention "uniform focal length." When the focal 
length of a lens is measured, I believe that the distance is usually 
taken from the center of the lens, but for any flat lens, the distance 
from the focal point to the edge will always be greater. Normally, we 
ignore this as the lens diameter is a relatively small proportion of the 
focal length, but for large lenses with very short focal lengths (f/D of 
<1, particularly) this can be an issue: - On receive, the incident angle 
may be great enough (e.g. "off to the side) that the arriving light may 
be blocked by the photodiode's housing. Also, the response of a 
photodiode is lower for off-angle light. - For transmit, taking a 
hypothetical case using an isotropic light source, a flat lens will not 
be evenly illuminated. For a large, short-focal length lens (low f/D 
ratio) the distance of the light source to the edges of the lens may be 
significantly greater that the distance of the light source to the 
center of the lens. What this means is that the entire surface of the 
lens would not be evenly illuminated - something that also means that 
the efficacy of the large size of the lens would be reduced as the 
smaller center of the lens would be doing most of the work. Note that 
with special optics, this last problem can be mitigated. 73, Clint 
KA7OEI Tom wrote:
> The detector is a PIN; the source is, at the moment, a 5mW 635nm laser, 
> visible on top of the buckets.  The initial application is rangefinding, 
> but I'll be communicating with them soon, I'm certain.  I've found some 
> ranging success with a simple TOF optical oscillator.  You're right that 
> the direct or mirrored beam would be too small to illuminate the entire 
> Fresnel, but the return is actually from a glass-beaded old stop sign 
> mounted in a tree 100m away, so the retro return is much larger than the 
> receive lens diameter and should be uniform.
> If I put myself at the focus and look toward the source-illuminated 
> Fresnel, it seems to me that, ideally, the entire surface of the lens 
> should yield light that is visible from the point of focus - that the 
> lens area is fully used to collect and focus all the energy it can on 
> that point via the detector's cone of acceptance, matching the full lens 
> area.
>
> I'm trying to collect the most sharp, brightest spot on the sensor, and 
> understand why that isn't perfectly possible.  The camera is a 
> diagnostic device to see what the lens is doing.  I now think I 
> understand that the focal length of this lens is not uniform across it's 
> diameter; I have a higher-quality Fresnel with a similar FL that I'll 
> try next.  Do you agree that - ideally - the entire Fresnel surface 
> should be illuminating the focus?
> Tom Cape Coral