K100D focus screen change

Hi,

I have found a few threads about focus screen changes, shims, front/back focussing etc.

I have just cut down a P30 screen and fitted it into my K100D. Focus with the screen is slightly off against the confirmation beep/green icon. It is very slight and outside I can just get the two to coincide.

I have taken some shots inside using my 50mm 1.7.

http://www.flickr.com/photos/40937348@N06/4490594948/in/set-72157623768488630/



As you can see the focus screen is closer than the icon/beep. I was focussing where the corner of the paper square meets the grid, centre of shot.

Are there any better tests I can do?...this seems to be the standard test to do.

Out of interest I have just put the old screen in and taken another shot using the focus beep...




Any comments/ideas please. It seems with this test my split focus screen in better 'tuned' than the camera's beep. Outside I was losing light and couldn't see much difference on the shots I took.

Cheers

Andy

ps..first attachments..fingers crossed....not showing in preview...adding anyway...ok...no pics!. Deleted three links and added link to the flikr set named 'test'

I have seen the recent debug mode thread, and seen how to do it on my K100d. When I get to the menu there is no AF adjust present. Is this the firmware?..do I need to update it or am I missing something.

When I go into debug mode the first screen shows Ver 1.0

Thanks

Andy

Andy I believe you need a Russian hack to get the Debug menu working but I could be wrong.

Are you certain that a P30 screen is the same thickness as a current one?

If it thinner then it would require shims to sit in the correct position.

Stefan, thanks, I will have a google..

Ken, thanks also, I will look out my micrometer. They are 'visually' the same thickness.

Andy

I have never understood why the thickness of the focusing screen has any effect on the focusing accuracy.

As I understand it the focused image is formed on the frosted surface of the screen and this is on the top side of the screen which if fitted with the original spacer between it and the penta chamber will be in exactly the same position as the original screens frosted surface and registration distance.

The only difference which I can see is that a thicker screen will make a slight difference to the pressure which the cradle frame will apply to the bottom of the screen once clipped into position but that will not change the focus registration distance.

Where am I going wrong

Greg

If the screen is thinner it wont sit against the prism, it will be lower.

Hi Ken,

I'm afraid I do not understand, the frosted top surface of the screen is forced up against the base of the prism by the carrier frame, if the surfaces touch or are separated by the spacer how can the thickness prevent it "sitting against the prism" even if the screen was 5mm thick the top frosted surface would still be in the same position relative to the prism as the original screens top frosted surface was. If the frosting was on the bottom of the screen then thickness would have an effect

I do appreciate that the screens position relative to the prism is not relevant to the registration measurement for focusing, the distance to the frosted surface of the screen must be exactly the same as the distance to the sensor surface.

Greg

Ken, the P30 screen is 5 thousandths of an inch thinner, which may or may not make any difference. Can't find any 'post it' notes anywhere to measure one of them, but anyway, as it seems better with the focus screen than the beep...I'll carry on experimenting.

Anyway, as my focussing with the P30 screen seems spot on, and the focus beep/auto focus seems slightly off, should I be concentrating on that and take it that the focus screen is ok?

Is close focussing with a wide aperture the best way of testing the focussing/autofocus?, is it the same throughout the range, or should I test at different distances with different lenses?

Also, I can get into debug mode but there is no AF adjust option, the ''Russian hack'' seems to be for later versions that locked the debug mode out...can anyone clarify?

At the moment I don't trust either way of focussing!


Thanks for the comments.

Andy

Andy