mooge
Well-known
Hardcore M2 Rangefinder Alignment
by Eugene J Lee / mooge
10-29-10
Purpose:
To demonstrate how to re-align and re-calibrate a Leica M2 or similar rangefinder unit that is beyond conventional means of calibration.
WARNING:
This is all delicate stuff. Be gentle. Place optics on a soft surface.
You're bound to get fingerprints everywhere unless you're really careful or wearing gloves, I'd actually suggest wearing gloves. The RF glass is something you don't want to clean; the more I cleaned mine, the uglier it got.
Theory:
The beamsplitter prism is our primary concern in a hardcore rangefinder re-alignment. The reason why the rangefinder is so out of whack is because the beamsplitter has shifted, and a small shift in the beamsplitter position can result in a large change in the RF image position. The beamsplitter prism can be shifted up and down, and rotated to a small degree along an axle in the RF unit frame.
The beamsplitter prism is secured with two flatsprings two grub screws, and an axle. There is a forward flat spring located towards the front of the beamsplitter, and there is a rear flatspring at the viewing end of the beamsplitter, just behind the rectangular viewfinder mask. Both flatsprings should be loosened if the beamsplitter is to be rotated.
The two grubscrews are tiny screws on the top of the RF frame that secure the beamsplitter in place. They can be used to shift the beamsplitter up and down, to adjust the vertical alignment of the RF image. The grubscrews should not be overtightened, for that may crack the beamsplitter.
The beamsplitter is secured with an axle thingy (I dunno what else to call it!), which consists of two metal poles that stick out from the beamsplitter prisms to fit in holes in the RF frame. While this axle does not go through the beamsplitter, it does not seem to be easily removable, which means that when the grubscrews are loose and the flatsprings are undone, the beamsplitter prism cannot be removed, only rotated on its axle by about 5 degrees. This rotation is used to put the horizontal part of the RF image into an acceptable range.
The concave lens is held in place with a spring clip secured by a screw, and a flatspting on top of the lens that pushes it down. Below the concave lens, there are two grubscrews that can be used to move the concave lens up and down, which also affects the vertical alignment.
When you remove this lens, you can look through the RF unit in all its 1:1 glory. It's really cool until you realize that the RF image and framelines are horribly put of focus.
R2 by rokka~!, on Flickr
R3 by rokka~!, on Flickr
R4 by rokka~!, on Flickr
This is a quick-shot of the note I made, mostly here for the diagram of the RF unit and its optics.
I'm actually missing an element (the concave lens between the beamsplitter and framelines mask is actually two elements, and not entirely concave), but the diagram is roughly correct.
Here is a correct diagram from a Leica brochure (if I recall correctly)
Procedure:
YES, the whole beamsplitter adjustment is really trial-and-error. And it takes a while to get in spec- the positioning is really finicky. The M1 isn't cheaper than a M2 because it's missing some little bits of glass, it's cheaper because nothing much needs to be adjusted.
You might have noticed that the concave lens is not mentioned in the procedure- I forget what I did with it. It does adjust the vertical a little...
You might wonder if the framelines are accurate after all this. I know I did. And the answer is... I dunno! You can check with a tripod and groundglass...
The Big Issue:
So, you're just about done and you've done the infinity, and just finished the 1m. All great... but for me, my infinity was off. I dunno if this is normal-but if the RF sensor arm is perpendicular to the lens axis at infinity, the 1m eccentric should NOT affect the infinity alignment. So I was pretty freaked out. In the end I adjusted both eccentrics until 1m AND infinity aligned, but what about the middle ranges? I don't know! That's the freaky part!
Take a look at the RF sensor arm- see the hole and the two notches? According to the KS-15(4) (the M2-R) repair guide, those are for some bending tool. Maybe this bending tool has something to do with the Big Issue. Actually, I just re-read a bit of the manual, and I think it might have more to do with making sure the RF roller is level with the RF cam of a lens. Hmm.
Big Issue Cont'd:
As normclarke will acknowledge below (I am... from the future!), after the beamsplitter shifts, the framelines will probably be out of whack. This can be changed by moving the convex lens between the beamsplitter and the framelines mask, and the framelines mask itself. The adjustment of the framelines mask is covered within the KS-15 repair manual below, and the convex lens can be loosened by loosening two screws on the back of the RF unit, beside the rear flatspring. You'll see it. Moving this convex lens will affect the RF image position.
The stand that normclarke mentions is pictured in the repair manual. Too bad I don't have one.
Dust on the RF patch?
If you look through the VF and see dust in focus on the RF patch or framelines, it's on the frameline mask. Brush off the little lens in the middle of the framelines, that should clear it.
Commentary:
Leicas are tough, everyone knows that! I slipped and hit my M2 against a rock on a camping trip, and when I told my friend about it, he said that it should be fine (it wasn't). Andrew Nemeth's Leica FAQ has an anecdote somewhere that claims that Leica would demonstrate the toughness of their cameras by throwing one against a brick wall and proving that it works fine. That's either rubbish or referring to SLRs- it only took me a ~60cm drop onto a moderately carpeted floor to put my RF so off that infinity aligned at 1 m.
What did I learn? Leicas are NOT that tough. Take care of your stuff. And RF alignment is a royal pain.
Final word:
I'm NOT trained or qualified in any way, and yes, I'm making this up- if this was covered in some sort of repair guide readily available, I wouldn't have made this write-up!
-----
Comments? Criticism? Answers to the Big Issue? Post 'em below!
Sorry if anything's missing or the writing is rough... I have to post this or else it'll never get out.
Bibliography:
M2-R service manual courtesy of Kim Coxon's Pentax Manuals. Most of this stuff isn't covered in this manual, that's why I wrote this up...
http://www.pentax-manuals.com/manuals/service
/leica_m2_service.pdf
RF lens diagram from Summilux.net
http://www.summilux.net/m_system/m2.html
by Eugene J Lee / mooge
10-29-10
Purpose:
To demonstrate how to re-align and re-calibrate a Leica M2 or similar rangefinder unit that is beyond conventional means of calibration.
WARNING:
This is all delicate stuff. Be gentle. Place optics on a soft surface.
You're bound to get fingerprints everywhere unless you're really careful or wearing gloves, I'd actually suggest wearing gloves. The RF glass is something you don't want to clean; the more I cleaned mine, the uglier it got.
Theory:
The beamsplitter prism is our primary concern in a hardcore rangefinder re-alignment. The reason why the rangefinder is so out of whack is because the beamsplitter has shifted, and a small shift in the beamsplitter position can result in a large change in the RF image position. The beamsplitter prism can be shifted up and down, and rotated to a small degree along an axle in the RF unit frame.
The beamsplitter prism is secured with two flatsprings two grub screws, and an axle. There is a forward flat spring located towards the front of the beamsplitter, and there is a rear flatspring at the viewing end of the beamsplitter, just behind the rectangular viewfinder mask. Both flatsprings should be loosened if the beamsplitter is to be rotated.
The two grubscrews are tiny screws on the top of the RF frame that secure the beamsplitter in place. They can be used to shift the beamsplitter up and down, to adjust the vertical alignment of the RF image. The grubscrews should not be overtightened, for that may crack the beamsplitter.
The beamsplitter is secured with an axle thingy (I dunno what else to call it!), which consists of two metal poles that stick out from the beamsplitter prisms to fit in holes in the RF frame. While this axle does not go through the beamsplitter, it does not seem to be easily removable, which means that when the grubscrews are loose and the flatsprings are undone, the beamsplitter prism cannot be removed, only rotated on its axle by about 5 degrees. This rotation is used to put the horizontal part of the RF image into an acceptable range.
The concave lens is held in place with a spring clip secured by a screw, and a flatspting on top of the lens that pushes it down. Below the concave lens, there are two grubscrews that can be used to move the concave lens up and down, which also affects the vertical alignment.
When you remove this lens, you can look through the RF unit in all its 1:1 glory. It's really cool until you realize that the RF image and framelines are horribly put of focus.
R2 by rokka~!, on Flickr
R3 by rokka~!, on Flickr
R4 by rokka~!, on Flickr
This is a quick-shot of the note I made, mostly here for the diagram of the RF unit and its optics.
I'm actually missing an element (the concave lens between the beamsplitter and framelines mask is actually two elements, and not entirely concave), but the diagram is roughly correct.
Here is a correct diagram from a Leica brochure (if I recall correctly)
Procedure:
- Top plate off. If you don't know how, you might be better off sending it to a pro.
- RF unit out.
- Turn the horizontal eccentric to its middle position, put the RF sensor arm back on.
- De-tension the flatsprings holding the beamsplitter in place, loosen the grubscrews.
- Futz with the beamsplitter.
- Tighten the grubscrews a bit so that the beamsplitter stays in place.
- Put the RF unit back in the top plate to check the okay-ness of it- if infinity is roughly when the RF sensor arm is parallel with the front edge of the RF unit (perpendicular to the lens axis), the prism is roughly in position. Keep an eye on the horizontal also- use the grubscrews to shift the beamsplitter up and down to correct this. It doesn't need to be perfect, just within the range of the eccentric.
- When it's all okay (cause okay is good enough!), tighten the grubscrews a bit more, don't tighten them too much- that might crack the prism. Re-tension the flatspring.
- RF unit back on.
- Top plate back on.
- Regular RF alignment. Infinity, then 1m; groundglass and loupe. Use a lens you know works fine, the faster/longer the better.
- Tighten 'er up, load with your favourite film, get out there. Dev, scan or print, share.
YES, the whole beamsplitter adjustment is really trial-and-error. And it takes a while to get in spec- the positioning is really finicky. The M1 isn't cheaper than a M2 because it's missing some little bits of glass, it's cheaper because nothing much needs to be adjusted.
You might have noticed that the concave lens is not mentioned in the procedure- I forget what I did with it. It does adjust the vertical a little...
You might wonder if the framelines are accurate after all this. I know I did. And the answer is... I dunno! You can check with a tripod and groundglass...
The Big Issue:
So, you're just about done and you've done the infinity, and just finished the 1m. All great... but for me, my infinity was off. I dunno if this is normal-but if the RF sensor arm is perpendicular to the lens axis at infinity, the 1m eccentric should NOT affect the infinity alignment. So I was pretty freaked out. In the end I adjusted both eccentrics until 1m AND infinity aligned, but what about the middle ranges? I don't know! That's the freaky part!
Take a look at the RF sensor arm- see the hole and the two notches? According to the KS-15(4) (the M2-R) repair guide, those are for some bending tool. Maybe this bending tool has something to do with the Big Issue. Actually, I just re-read a bit of the manual, and I think it might have more to do with making sure the RF roller is level with the RF cam of a lens. Hmm.
Big Issue Cont'd:
As normclarke will acknowledge below (I am... from the future!), after the beamsplitter shifts, the framelines will probably be out of whack. This can be changed by moving the convex lens between the beamsplitter and the framelines mask, and the framelines mask itself. The adjustment of the framelines mask is covered within the KS-15 repair manual below, and the convex lens can be loosened by loosening two screws on the back of the RF unit, beside the rear flatspring. You'll see it. Moving this convex lens will affect the RF image position.
The stand that normclarke mentions is pictured in the repair manual. Too bad I don't have one.
Dust on the RF patch?
If you look through the VF and see dust in focus on the RF patch or framelines, it's on the frameline mask. Brush off the little lens in the middle of the framelines, that should clear it.
Commentary:
Leicas are tough, everyone knows that! I slipped and hit my M2 against a rock on a camping trip, and when I told my friend about it, he said that it should be fine (it wasn't). Andrew Nemeth's Leica FAQ has an anecdote somewhere that claims that Leica would demonstrate the toughness of their cameras by throwing one against a brick wall and proving that it works fine. That's either rubbish or referring to SLRs- it only took me a ~60cm drop onto a moderately carpeted floor to put my RF so off that infinity aligned at 1 m.
What did I learn? Leicas are NOT that tough. Take care of your stuff. And RF alignment is a royal pain.
Final word:
I'm NOT trained or qualified in any way, and yes, I'm making this up- if this was covered in some sort of repair guide readily available, I wouldn't have made this write-up!
-----
Comments? Criticism? Answers to the Big Issue? Post 'em below!
Sorry if anything's missing or the writing is rough... I have to post this or else it'll never get out.
Bibliography:
M2-R service manual courtesy of Kim Coxon's Pentax Manuals. Most of this stuff isn't covered in this manual, that's why I wrote this up...
http://www.pentax-manuals.com/manuals/service
/leica_m2_service.pdf
RF lens diagram from Summilux.net
http://www.summilux.net/m_system/m2.html
Last edited:
Vickko
Veteran
Wow. Congratulations.
axiom
Non-Registered User
Valuable info.
Bookmarked
Bookmarked
Luna
Well-known
He speaks the truth!
Great read. And this link explains how to remove the top plate off pretty clearly. ( http://www.jumboprawn.net/jesse/cams/gear-profiles/leica-m2-overhaul/leica-m2-overhaul.html )
normclarke
normclarke
Nice write up but I believe it is more complex than you show. To complete the alignment needs the double lens that sits in between the frame lines and prism to be considered along with the frame lines and the field lens. There is a lot of adjustment of all these elements before the job can be considered finished. It not only concerns focus at infinity and 1mtr but frame line accuracy with respect to the final image. I think this is one of the areas of Leica repair that does not get covered in most available info on the web. This is one of the secrets that the real pro's do not advertise! Special jigs are used
for this alignment.
Best,
normclarke.
for this alignment.
Best,
normclarke.
_goodtimez
Well-known
That kind of technical email makes my day ! Thanks a lot for sharing such good food.
mooge
Well-known
thanks guys!
Normclarke- you're right about that frameline stuff. I've done an edit now... and I'll check the frameline accuracy tomorrow.
Y'know what? there was that condenser lens that they removed mid way through the M4-2 line, right? Does anyone know which lens it is? I can't imagine the RF working if they took anything out!
cheers.
Normclarke- you're right about that frameline stuff. I've done an edit now... and I'll check the frameline accuracy tomorrow.
Y'know what? there was that condenser lens that they removed mid way through the M4-2 line, right? Does anyone know which lens it is? I can't imagine the RF working if they took anything out!
cheers.
normclarke
normclarke
The lens is the one that sits on the side of the frame line mask that faces the small lens that forms the image of the moving image derived from the pencil prism.
Best,
normclarke.
Best,
normclarke.
Mister E
Well-known
Awesome writeup.
mooge
Well-known
another edit...
Normclarke, I didn't even notice that lens... it's actually missing from my diagram! how'd you learn this stuff?
thanks Nate. your Central Asia series is awesome.
and I just checked, my framelines are okay. yay.
Normclarke, I didn't even notice that lens... it's actually missing from my diagram! how'd you learn this stuff?
thanks Nate. your Central Asia series is awesome.
and I just checked, my framelines are okay. yay.
normclarke
normclarke
With many years experience of working on mechanisms and optics in the space industry.
Best,
normclarke.
Best,
normclarke.
goat2017
Newbie
mooge:
I don't know M2 but on M3 there is a "hidden" screw to adjust the frameline mask position (diagonally). Once the horizontal viewport is accurate, adjust the two grub screws holding the convex lens next to the main prism for vertical adjustment. This will require disassembling of the whole rangefinder unit. Leica M frameline is not accurate as an SLR. You may accurately adjust the frameline for 1m, but at infinity the actual viewport captured by the lens/film is larger than the frameline shows. And you can always crop.
I don't know M2 but on M3 there is a "hidden" screw to adjust the frameline mask position (diagonally). Once the horizontal viewport is accurate, adjust the two grub screws holding the convex lens next to the main prism for vertical adjustment. This will require disassembling of the whole rangefinder unit. Leica M frameline is not accurate as an SLR. You may accurately adjust the frameline for 1m, but at infinity the actual viewport captured by the lens/film is larger than the frameline shows. And you can always crop.