RobertOToole wrote:It appears that really well corrected lenses, like a Macro Varon, work wonderful on extension but they really badly on a tube lens. On the other hand lenses that are okay or just average on extension, like a Componon 4/28 or 4/35, put these on a good tube lens and they work beautifully on a tube
Hi Robert, doesn't this follow almost by definition? Macro and scanner lenses are optimized for use at a short finite - short finite conjugate ratio, enlarger lenses are optimized for short finite - long finite, and tube lenses are intended for infinite - finite use. The long finite (print) side of an enlarger lens is presumably going to match a tube lens focused at infinity better than a short finite design.
A testable prediction from the conjugate ratios is, if the tube lens can accommodate it, focusing the tube lens at distances closer than infinity should improve performance when the other lens isn't infinity corrected. With enlarger lenses the best tube focus distance would presumably be the one corresponding to the manufacturer's preferred magnification for the enlarger lens (d = M f, often around 400-600 mm). However, since performance is already good at infinity, the potential for improvement seems likely to be small. This is probably a way of saying a conjugate distance of several hundred mm isn't greatly different from an infinite one in terms of ray angles.
A macro lens at close focus is perhaps a natural close finite match for a scanner lens on distance though, as Rik points out below, ray angles are subject to increasing mismatch as magnification increases. I have neither a scanner lens or a tube lens to test with but this also may not be suitable for epi illumination as working distances are probably quite short.
mawyatt wrote:The Raynox and Thor Labs are simple lens, so probably not "well corrected", but make superb tube lenses.
Hi Mike, well corrected should be fine so long as it's well corrected for the right things. Since Raynox DCRs are intended for use as close up lenses on telephotos with limited close focus distances they should be optimized for providing a long to infinite conjugate to the telephoto when the subject plane is near their focal length. Reversed, that makes them much like a tube lens.
Thorlabs doesn't appear to disclose their tube lens design---the Zemax models are indicated as black box---but for what they charge for their lenses I would hope they're well designed and manufactured for their purpose, which seems consistent with Robert's overall best rating in his
tube lens test. From the dimension drawings I'm unsure if double Gauss or similar designs can be excluded.
A search is giving me lots of hits repeating the 135 f/2 complaint but not posts with test results supporting that conclusion. I suspect Rik may have already indicated the difficulty in other contexts but am hesitant to suggest a linkage without direct supporting evidence. Indirect evidence (some which comes from your observations of using other 135 mm lenses as tube lenses
) does admit interpretations where level of correction isn't the trouble.
Scarodactyl wrote:I would imagine the same is true with microscope tube lenses (ie, using a nikon objective on a zeiss scope).
I don't see why it wouldn't be. From a practical standpoint the trouble is that, unless one's designing the whole system, it's unlikely to be able to match compensating aberrations well enough to produce results competitive with those from independent correction.
