Chiopt 3.5x Line Scan lens

[ray_parkhurst]

I found this lens in my search for 3-3.5x lenses for sensor-pan stack-and-stitch work. The company is relatively new (2010) and this lens is even newer, so does not seem to be available yet on surplus market. It has impressive specs, 82mm image circle, nominal aperture f1.6. This large aperture gives hope of exceeding the performance of the RayFact 3.5x.

I contacted Chiopt for a quote, and it came back at US$7k for 1 or US$6250 for 10 pieces. Expensive, but cheaper than the Nikon. I asked for a review sample but was turned down unfortunately.

Anyway, here's another lens to look for...

http://www.chiopt.net/product/90.html

[ChrisR]

Resolution 1µm. simply saying that takes one pixel, leads to 5.3 Gpixels per view.

[ray_parkhurst]

I think they mean on subject. Jibes with the 3.5x mag and "3.5um cameras" resolution.

[rjlittlefield]

I calculate that nominal f/1.6 at 3.5X gives NA 0.243 on subject. At lambda 0.55 microns that gives diffraction-limited cutoff at 0.884 cycles per micron, with MTF 32% at 0.5 cycles/micron (Nyquist at 1 micron per pixel).

That all sounds like a good spec to me.

--Rik

[Scarodactyl]

With that kind of coverage and resolution it sounds like a great cmo for a stereoscope or macroscope. Or rather it would be much better since the resolution is edge to edge, not just at the center. Of course you'd need a zooming tube lens that didn't stop down at lower magnifications. Hmm.... I guess I'll push that idea to the back of the shelf for now.

[ray_parkhurst]

Now someone just needs to get hold of one to see if it lives up to the expectations.

[ray_parkhurst]

I calculate that nominal f/1.6 at 3.5X gives NA 0.243 on subject. At lambda 0.55 microns that gives diffraction-limited cutoff at 0.884 cycles per micron, with MTF 32% at 0.5 cycles/micron (Nyquist at 1 micron per pixel).

That all sounds like a good spec to me.

--Rik

Rik...since they spec the lens for a 3.5um pitch sensor, it does not make sense that the 1um spec would be for image resolution, even if the lens can theoretically achieve it. I'm pretty sure they are talking about system resolution (limited by the sensor), not lens.

If the lens is truly diffraction-limited at f1.6, and could actually achieve 32% MTF at 1um pitch, well that would be quite a feat at 3.5x across an 82mm image circle! It's of course what I was hoping for when I started the thread. That would be f7.2 effective at 3.5x!

I see little if any improvement in performance from the RayFact 3.5x going from f2.8 to f2.4, so I would conclude the lens is not diffraction limited at f2.4. Maybe it is at f2.8, I'm not sure. My expectation of the Chiopt is that it would require stopping down a bit to achieve max sharpness, but how far? If only to f2, then it's awesome! If it needs f2.8, then it would be similar to the RayFact.

[rjlittlefield]

I calculate that nominal f/1.6 at 3.5X gives NA 0.243 on subject. At lambda 0.55 microns that gives diffraction-limited cutoff at 0.884 cycles per micron, with MTF 32% at 0.5 cycles/micron (Nyquist at 1 micron per pixel).

Rik...since they spec the lens for a 3.5um pitch sensor, it does not make sense that the 1um spec would be for image resolution, even if the lens can theoretically achieve it. I'm pretty sure they are talking about system resolution (limited by the sensor), not lens.

I agree. Apparently what I wrote was too brief to be understood correctly. Do these added words help?

I calculate that nominal f/1.6 at 3.5X gives NA 0.243 on subject. At lambda 0.55 microns that gives diffraction-limited cutoff at 0.884 cycles per micron on subject, with MTF 32% at 0.5 cycles/micron on subject (Nyquist at 1 micron per pixel on subject = 3.5 micron pixels on sensor at 3.5X).

--Rik

[ray_parkhurst]

I calculate that nominal f/1.6 at 3.5X gives NA 0.243 on subject. At lambda 0.55 microns that gives diffraction-limited cutoff at 0.884 cycles per micron, with MTF 32% at 0.5 cycles/micron (Nyquist at 1 micron per pixel).

Rik...since they spec the lens for a 3.5um pitch sensor, it does not make sense that the 1um spec would be for image resolution, even if the lens can theoretically achieve it. I'm pretty sure they are talking about system resolution (limited by the sensor), not lens.

I agree. Apparently what I wrote was too brief to be understood correctly. Do these added words help?

I calculate that nominal f/1.6 at 3.5X gives NA 0.243 on subject. At lambda 0.55 microns that gives diffraction-limited cutoff at 0.884 cycles per micron on subject, with MTF 32% at 0.5 cycles/micron on subject (Nyquist at 1 micron per pixel on subject = 3.5 micron pixels on sensor at 3.5X).

--Rik

Yes, thanks for clarifying. All makes sense.