Quote:

---

Yes, thanks Glenn for an explanation we non-scientist/engineer types can comprehend. Just curious how that plays out then in a newtonian. So a 1/10 wave P-V primary really delivers 1/5 lamda to the secondary. If the secondary is say 1/20 wave, then what's the final lamda?

---

Yes, a 1/10 lambda *surface* deformation delivers a 1/5 lambda wavefront. In fabrication, testing of an optic is based on the wavefront sent to the test apparatus, and therefore takes into account this doubling of error. That's why one must ascertain just what the specified quality refers to; the surface or the wavefront.

In assessing cumulative errors, the standard method is to take the square of the sum of the squares. To work with your specified values:
1/20 = 0.05
1/5 = 0.2

resultant lambda = SQRT (lambda1^2 + lambda^2)
= SQRT (0.05^2 + >0.2^2)
= SQRT (0.0025 + 0.04)
= SQRT (0.0425)
= 0.206
~1/5 lmbda

As you can see, the worst optic is the weak link in the chain, because its waveform error so dominates the better optic's. If one mirror was perfect and the other had lambda 1/5 errors, it's easy to see that the resultant error will also be 1/5 lambda.

--------------------
Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV

Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.

Post Extras