2 element semi apo v 3 element apo

[adamsp123]

Just some thoughts on possibly why cheaper 3 element APO refractors apparently,so I hear, have more quality issues in comparison to 2 element semi APOs.

I went on a course years ago about the topic of tolerances and manufacturing problems.

One of the case studies was about a manufacturer of engine timing chains, the ones that never broke, not

like these silly rubber bands we have on our car engines today that snap and leave us with a massive

bill to pay.

So this manufacturer made millions of these things and just occasionally he would get a batch

that would not fit on the engine sprokets properly and although the individual parts of the chain

were within tolerance the suckers were no good, why?

Well it turned out if one of the parts making up the chain assembly was right at top tolerance and

a certain other part was right at bottom tolerance then the problem occured so... the answer

in this case was to make sure that the parts were sorted so this combination did not happen in

assembly.

Now to look at the tolerances comparing a 2 element semi apo refractor versus a 3 element apo we have

tolerances of the lens faces and postional tolerances of the glass elements to each other.

Taking a simplistic look at possible maximum and minimum tolerance combinations for a refractor objective, and as I am not a optic designer some assumptions may be wrong but I think the main argument should stand up.

So for a doublet we have each lens 2 faces that can be at top tolerance through to bottom tolerance

so for one lens we have 4 possible combinations of max/min tolerances, and combining the second

lens we have 4X4 possibilites = 16

We could then have postion tolerances of the lenses to each other once assembled in the OTA.

For example horizontal spacing/fit we have max or min distance so =2

Vertical position for each lens once in the tube 2X2=4 ie high/high high/low low/high low/low relative to

the eyepiece position.

Skewness, 2 for each lens=4 .There maybe many other factors but I'll keep to these.

So we have 16X2X4X4=312 possible top/bottom combinations.

Now for a triplet lens faces we have 4X4X4=64

horizontal spacing 2 for lens 1 to 2, 2 for lens 1 to 3, and 2 for lens 2 to 3 so 2X2X2=8

Verticals the same combinations as horizontal spacing =8

Skewness 2X2X2 =8

So a triplet is 64X8X8X8= 32768

Thats roughly 10 times the possible top/bottom tolerance situations in my simple scheme.

Now of course in most OTAs the parts will typically be somewhere in between but there is

ample more room for a situation, as in the engine chain example, that a certain combination

of lens tolerances makes the OTA poor and finding the cause can virtually impossible.

And so maybe thats why cheaper end triplets can have more issues and that is why the best of breed

triplet OTAs are so expensive, they will have much tighter tolerances and better knowledge and

manufacturing proceedures costing alot of time and hence money.

This is not to say that perfectly OK cheaper triplet OTAs are out there as they clearly are!

Hope I haven't made any glaring errors, no doubt I will be put right/slapped down if need be.

Pete

[Nostromo]

Hi there, here's an interesting article on triplet lens design by Thomas M Back giving a bit of background on the various challenges faced with triplets, and his own definition of what an apochromat is. Nick

[adamsp123]

Hi Nick,

Thanks for the link, it seems however you define the telescope there is a host of descriptions but at the end of the day it all hangs on how good it works to the user, be it an eye or a camera/CCD.

My topic was just trying to show how difficult it is to manufacture from a simple engineering point of view - let alone the black art of optics design, of which I only know the very basics.

pete