转自cloudynight2004年的帖子
Diagonals; Prisms vs. Mirrors
Neither mirror nor prism are better. If you notice, every manufacturer of quality scopes with fast focal ratios uses mirror diagonals. Every manufacturer that sells quality scopes with long focal ratios sells prism diagonals. It抯 not an accident. You get the diagonal that matches the scope. Assuming the same quality of manufacture, prisms give less light scatter, a good thing. And unlike a mirror, they are also self collimating. But they also create false color in scopes with fast focal ratios. Just like your old Jr. High experiments, when you send light into a prism at an angle other than perpendicular, it will cause false color. Fast telescopes send light into a diagonal prism at a steeper angle and can in fact introduce false color into a scopes where there was none. The sharper the angle (faster the telescope) the more false color is introduced. Prisms are more expensive to make in larger sizes than mirrors and may account why in 2?and larger diagonals, mirrors dominate. Prisms on the other hand never scatter light and therefore can produce excellent contrast. Also, they never age. In terms of transmission, each depends on the coatings. A multicoated prism and enhanced Aluminized mirror can both achieve about the same 98% transmission. But that really is not important. (WHAT DID HE SAY?) Most of the discussions about diagonals have centered on reflectivity. Of all the characteristics that make a good diagonal, reflectivity is one but not the most important. But it is the easiest characteristic to market, 96% is better than 95% right. There is a vast amount of laboratory data that proves that human beings cannot perceive a few % of light change, yet it is the characteristic most often quoted to define how good a diagonal is. Buy X because its 97% and not Y because its only 95%. Yet even Meade抯 and Celestron抯 new ads for their high transmission coatings state that you CAN NOT really 搒ee?the difference and they抮e talking about 10% difference. On the other hand, a camera can easily 搑ecord? the difference. But you don抰 use diagonals when photographing do you. So, why is it that people that demand full documentation on the optical accuracy of their $5000 telescope, only worry about the brightness of their diagonal and not accuracy? Isn抰 accuracy, in this case flatness, at least as important than brightness. Doesn抰 the difference between a 1/7th wave vs. a 1/20th wave surface have a greater impact on your viewing than the difference between 94% vs. 98% transmission. The few that supply accuracy papers on their diagonals do so before they are coated, not after and there can be a huge difference. The papers make you feel good, but do not reflect (sorry) the actual product you receive. As a rule, the brighter the coatings, the more layers of coatings had to be laid down, and the less it reflects (sorry again) the accuracy of the original precision flat surface accuracy. Standard Aluminum coatings (88%) have one layer, semi enhanced (91%) 2 layers, enhanced Aluminum (94%)has 3 layers of coatings. Dielectric coatings can have up to 22 layers. The advantage of a dielectric coated mirror is its longevity. Unless abused, they will last a lifetime. In short, use the diagonal that works best your scope. It may be the one the manufacturer supplied.
Look for the most accurate diagonal, then the brightest. Look for all metal parts, a machined housing is best but not required. A true 90.00 degrees is important. That means 1 piece CNC machined housing. User collimatable is nice. The substrate is important. Cervit, Zerodur, Astrosital and Quartz are better than Pyrex. Pyrex is a bit better than BK7. BK7 is better way than plate glass.
| 
增加点人气，翻译我放在自己博客里面了
置顶卡
变色卡
千斤顶
显身卡
