Leica IIIa Fast Shutter Speeds

I am the proud new owner of a IIIa. From the first batch, and alleged provenance includes being taken from a German tank officer (interesting and sad for him if true).

I've shot a few test rolls after an "amateur CLA." This includes removing the shutter crate, ronsonol to all of the ancient grease, and some fresh OTIS (dry grease) applied to the greased mechanisms. I love the images. I'm using the Elmar 50/3.5 and 90/4. There's that vintage look that I was after. However, the fast speeds of 1/100 and above are about 1-2 stops slow.

Reviewing the curtain gap with a backlight and iPhone shooting 240fps video, it seems like the gaps are large. 1/100 leaves a little more than half of the frame open. 1/200 about a quarter. 1/500 close to 1/8, and 1/1000 barely less.

Given a curtain transit time of 1/20 (which I've read elsewhere on the forums), I'm assuming I should be about 1/2 at 1/60 and roughly decreasing by halves thereafter down to about 1/48 (just shy of a millimeter) at 1/1000.

I know the sure fire bet is to send to YYE or another professional for a full CLA, but I haven't got 150 in the camera, and I'm comfortable shouldering some risk.

My main question is if I should try:

1. disassembling further for more CLA (I.e., disassembly of the shutter crate)

2. Reattaching current curtains at a different point (they are in good shape and appear original at least in the sense that they are sewn and not quite fresh)

3. Further disassembly and curtain replacement.

If I go 2 or 3, or even 1 and 2 or 3, I'm curious as to how I should vary current placement to close up the gap.

I have the utmost respect for this camera, the right toolkit, and experience on quite a few other disassmblies/CLA. I don't expect perfect 1/1000, but I'd like to shoot asa100 f16 1/100 on a sunny day and have faster speeds than my Argus C3.

Thanks in advance.

P.S. Included is a collage of 12 images. It is 1/30-1/1000. 6 opening and 6 closing.

 

"I'm curious as to how I should vary current placement to close up the gap" tells us that you don't know where you are headed and suggests that further disassembly will serve no purpose.

I'm not sure I understand your response.

I freely admit I don't know where to head next.

Hoping someone has been in this spot before and can give me some pointers. Other forum posts and Christopher Kuni's guide on in-camera replacement have said that, on the IIIa, curtain placement is the only way to adjust high speeds. I've not seen any more said about it than that though. It would seem that attaching the opening curtain "further back" on the drum is the necessary course (assuming the root cause doesn't sound like just more cleaning is needed).

I would not trust the iPhone video to evaluate shutter operation. Rather, I would shoot bracketed exposures and look at the negatives.

Fair point, Dan. My first test roll showed about +2 exposure at high shutter speeds. That's when I did a cleanup. Next roll I shot bracketed and was still more than +1 judging by the unadjusted curves in my scanner. I appreciate the point about the iPhone, and I won't obsess over the videos. I will focus on the quality of the negative. I don't shoot slides, and I shoot sunny f16 99% of the time. My goal is to get reasonably accurate high speeds...at least up to 1/500 and hopefully something recoverable on B&W negatives at 1/1000.

Thanks.

What I mean is that you are "flying blind". Unless there is a response here from someone who knows, or unless you find reliable information elsewhere, your ability to fix the problem is uncertain. [edit] If curtain placement was correct, you should get correct speeds unless the mechanism is dirty or unless a component has failed. I think more thorough cleaning and lubrication are called for. Edited March 11, 2017 by mukul_dube

Thanks, Mukul. I'm leery of getting into the shutter crate, but I really love the feel and experience of this camera.

I think I will run another roll or two through and see how inconvenient the missing high speeds really are. After all, it's a working camera right now.

I am open to more feedback from anyone, but I think I'll try and adjust to the camera for right now, and go for a deep cleaning if I need more.

Thanks again.

OK, sounds like the shutter needs some TLC. I'd have to send it to DAG, or some other service provider. I had my IIIf red-dial CLA'd by YYE a couple years ago, and it's shutter has been running accurate ever since, even on the low-speed dial. But, you're looking at a couple hundred $$.

The worst thing is I know you're right, Dan. I just can't swing the CLA right now. I got the camera on the cheap. And while I know you should budget the CLA with a screw mount (and most vintage rigs for that matter), the only way I could afford one was on the cheap and taking a gamble that I could get it to usable.

You could compensate for the shutter error with the aperture.

The position of the curtains on the drums and rollers was set correctly at the factory, and should never need to be changed. They used jigs. If the exposure is uneven from one end to the other of the frame, either the bearings on the drums and rollers are gummed up, or the curtain tensions are wrong. You need a optical shutter speed tester to adjust the tensions.

Quick update:

I shot another roll bracketed. It was all about 1-2 stops slow, and worst of all capping at 1/500. Which means that I lost well exposed shots to capping shooting sunny f16 with asa 100.

Mukul: fortunately my Elmar 50/3.5, Elmar 90/4 and Jupiter 12 35/2.8 all stop down to f22 or better. Adding your advice to the bracketing on my last roll salvaged some pics. Thank you.

So, being the impatient, desperate, infatuated owner that I am, I went ahead with a shutter crate disassembly. The process is straightforward. Fortunately the Russians were meticulous enough in their copying that the instructions at fedka for a zorki were spot on

http://www.jay.fedka.com/index_files/Page455.htm

Two primary difficulties: disassembly of an 80 year old camera is not as simple as knowing which screws to turn in which order...a couple of parts were frozen, and for those of us born with only two hands it takes as much luck as finesse to reassemble. It was a pain to reorient the drum. The first reassembly, I was off by one tooth (advance to drum) which resulted in the second curtain not latching open. Of course it is several steps of assembly to test and several steps of disassembly to correct followed by several steps of reassembly to test again.

Based on John's feedback, I did not mess with the curtains, but I focused on the contact points of the drum and rollers. Cleaning with Ronsonol and applying a very light oil (non-responsive yoyo bearing lube).

Then, KINDA following what Dan said, I reshot high speed iPhone video with a healthy dose of skepticism to adjust my curtain tension.

I think it's possible that whatever error is present in the video is constant and repeatable. I know from using the iPhone on my Salyut that what looks like capping on the phone will definitely cap, but sometimes Half of the negative will be severely underexposed where the phone video does not suggest it. This makes me think that there is certainly error in the video, but that it may be consistent and I can account for it. In the video, it appears that the curtains speed up as they transit the aperture. In which case, it makes sense that I would want the iPhone video to look like the curtain gap is increasing. Whether or not this is reality, I do not know.

The camera definitely feels better and the exposure looks better after the crate disassembly. I have readjusted curtain tension based on this logic and loaded a roll of Delta 100 (because it is relatively less tolerant of exposure error...and I like it) to shoot some more bracketed test shots.

I will update the post based on the results, but will also say that if they are not satisfactory I will probably be parting with some of my other equipment to try and finance a trip to Youxin.

Looking at all the work you have put in, I find myself hoping that you won't need Y. Ye.

Definitely what John S. stated.

"Reattaching current curtains at a different point" Matthew V.

Unfortunately this thought process is tantamount to correcting the timing of an untampered with old school engine, by jumping crankcase gear teeth instead of simply moving/adjusting the ignition distributor assembly...

I can appreciate that. Brainstorming gone wild. Thanks to the advice of the group, I'm sitting here a week after my initial post VERY glad that I have not violated the curtains.

Gus, mad respect to you, sir. Thanks for stopping by. Reminds me of getting autographs when I was a kid.

The Delta 100 is developed.

And the results are encouraging.

Took about six pictures bracketed 1/60-1/1000 on each shot. My stupid hat hit the shutter speed dial on a few. I'll chalk it up to operator error/the "difficulties" of owning a LTM.

I loved this camera, and it was smooth after CLA one. But after shutter crate disassembly and clean/lube smooth has a whole new meaning. Wow! One of my coworkers today told me that when he was a kid in Central Kentucky Leicas were the only cameras allowed in the courtroom because they were so quiet. I assume this is true, but it's certainly sensical based on what I hear/feel now.

Great news: the overexposure is gone. And I got some decent pictures where they should be within the bracket.

Bad news: there is capping at 1/200(slight)-1/1000 (2/3 of the frame missing).

Based on the fact that the left side of the image (shutter opening side) is well exposed and the right side (shutter closing) is underexposed, I have loosened the closing curtain as opposed to tensioning the open. . Seems correct, but please someone tell me if this is illogical.

Went two teeth of the curtain tension gear and loaded back up with some Kentmere 400.

Will advise again after this roll.

Thanks again to all for your responses and a million thanks to everyone who did not suggest I mess with the curtains.

Developed the K400. Results are improving. Good exposure and amazingly reduced capping.

The positives are good without any adjustment. In the negatives, I still see some lightening on the right side of 1/500 and 1/1000. I've given one more sprocket of loosening on the second curtain.

I've loaded a roll of Tri-X 400 and will post results. It slightly bothers me to add one more variable, but this is how I shoot. After 187 rolls, I still haven't quite made my mind up on films

Forecast shifted to sunny so I've Swapped the Tri-X for FP4+. Haven't finished the roll yet though.

Updating the thread regarding iPhone slo-mo for shutter testing. Probably should've done some math in advance of using it.

Given a curtain transit time of 1/20, and an aperture of 36mm, this suggests that the curtains are moving at 36/(1/20)=720mm/sec. Then we consider 240fps video speed. This means 720/240 mm/frame=3mm/frame.

If 1/20 shutter speed is full aperture, then 1/500 should be 1/25th of the aperture or 1.44mm.

This means that the iPhone is trying to see a 1.44mm gap when each frame the curtains should be moving 3mm. Even worse at 1/1000 when you should have a .72mm gap.

Just more confirmation that Dan was correct in disregarding the video.

 

I have always believe the shutter time is 1/60 for the Canon and Leica rangefinders, though maybe not for the earliest models.

 

But if you can measure the opening with your 240FPS video, you should also measure the curtain transit time.

Glen: I think this is one of the main arguments that people provide in favor of iiiC over iiiF: very slow flash sync as a result of the CTT.

I've been hesitant to update because I keep thinking "one more roll." Knowing that my camera was thoroughly cleaned down to the spindles, I started adjusting curtain tensions to chase capping observed in test rolls...and then overexposure. I was getting to the point that I had way too many turns on the tension screws. Then I remembered something that I had read in a forum about "shimming" the second curtain ribbon with masking tape at the base layer. Upon further inspection, I think my curtain ribbons have been monkeyed with because I see no spindle in any example pics, but I can see the bare spindle/end of the ribbon when my shutter is tensioned. So I gave some backwards turn to the take up side of the second curtain and put three layers of masking tape the same width of the curtain. Test roll showed capping.

So I let all of the tension off and started over with tensioning. At this point, I returned to the iPhone. I surmise it's not as simple as having an x-mm gap across. Just like the official drum tester, it should be a little bit of craft steeped in math.

Results look good (I.e., gap widens as curtain transits based on the idea that the spindles are growing in circumference). Tested with some Instax film: Transferred from pack to Leica film gate in dark bag. Removed Leica. Exposed. Returned to bag, slid shot back into pack, and shot instax camera with hand over lens in bathroom with door closed and lights out.

Have loaded up another test roll and will update after developing.

Maybe it is now a lost art, but the way we did shutter speed testing 50 years ago was to take pictures of a TV screen.

 

For analog TV, the scan rate is very accurately determined at the transmitter, with the line rate at 15734 Hz. You get a diagonal stripe from a focal plane shutter, but if you count the scan lines along a vertical line, and divide by 15734, you get pretty close. The angle of the diagonal stripe gives the curtain travel time, and varying height along the stripe allows measuring unequal curtain rates.

 

But for digital TV sets, the timing is all determined locally. Even so, it should be pretty close with a crystal controlled oscillator, though finding the actual frequency might not be so easy. A CRT set run off a conversion box, or from a VCR, should still follow the old timing. LCD sets don't scan the same way, so might not work as well.

I didn't know about the IIIc. From about when I was born, my dad had a Leica If and Canon VI. I now have those. From when I was pretty young, I knew about the 1/60 curtain time. Recently, I got a Mercury II, with a rotating solid metal focal plane shutter. That one I believe is closer to 1/20 of a second. But I didn't know any Leica's were that slow.

My God, man, your tinkering - the cost of your time & test rolls/development will probably exceed Youxin's fee for doing the job right. Just send it to him and have a camera you can enjoy for many years to come. You may have gotten the camera on the cheap, but you can also severely damage it on the cheap.

The IIIc doesn't normally have flash sync at all, though some were later upgraded to add it, making a camera with similar specs to the IIIf - do these have different shutter mechanisms to an original IIIf?

"do these have different shutter mechanisms to an original IIIf?" Richard W.

No, they're not appreciably different from the last model (IIIc) prior to the Leica IIIf BD.

The 'new model' IIIf, added a cleverly designed speed dial and top-plate that accommodated the flash needs of that era.

But this 'piggy back' (add-on) mod, did add to the complexity and later to *performance ageing issues.

*A service C L A became very critical to sustaining proper high speed tolerances...