Contax AX/RX mirror slip repair and adhesive replacement - how to determine carrier adhesive thickness?

[Ziotoo4]

Hi all,
I recently purchased a contax AX to use with my C/Y lenses.

When the camera arrived I checked for signs of repair of the dreaded mirror slip and sure enough there was some evidence of work done.

The autofocus (done via a CMOS at the bottom of the mirror box) and the focusing split prism on the focusing screen agreed with each other, but when focusing the lens at infinity (infinity stop), the split prism and the autofocus sensor seem to think the object is around 20m away.

So I shot a test roll with targets and sure enough all the images are backfocused:

* If I focus on an object "20m away" according to the split prism or the autofocus, the object is actually at infinity.
* If I focus on an object "2.5m away", the lens distance scale is at 3m.

The lens is accurate on my other contax bodies.

After making sure the camera had never been opened before (factory thread lock is still in place), I concluded this was a botched mirror reattachment job.

So I spent some time with acetone, ethanol and carefully loosened the adhesive used. With a very thin blade I slowly cut the mirror away from the carrier.

Sure enough, indeed someone did some double tape magic here in the past. I very carefully removed the goop from the carrier (taking care to remove the focusing screen first and not ruin the black flocking paint with the solvent), and also cleaned the back of the mirror with ethanol and patience, taking care not to scratch the front.

Now both are clean, I need to reattach the mirror onto the carrier. How do I determine the thickness of the adhesive to be used so that the focus at infinity is correct?

I could do by trial and error with different thickness tape, but:
- I know an equation must exist to calculate the flange to focus screen delta I need to apply to compensate for the incorrect distance to the object I am focusing on
- contax must have used standard thickness adhesive, since the repair manual show that focus adjustment on these cameras requires opening the whole body and shimming the focus screen + reprogramming the ROM, which is super time intensive - so I can't imagine this to be a process for every body leaving the factory.

Thank you so much in advance, and for helping me give this old camera a new life!

 

[MattKing]

Welcome to Photrio.
I've moved your thread to the Camera Repair sub-forum.
We hope you find the information you need.

 

[Ziotoo4]

Thank you Matt! Apologies for missing the repair forum. Thanks for the warm welcome!

 

[reddesert]

The sign of the focus offset is still a little unclear to me - just a phrasing issue. Let us assume that the lens focusing scale is correct. Are you saying that when you point the camera at an object that is physically at infinity, the focusing screen indicates correct focus when the lens is moved to 20m on the focusing scale? Or the other way around?

If you tell us the focal length of the lens you're using, we can calculate the depth offset (the mirror would be offset by half the amount of lens extension between infinity and 20m). The formula is that
1/f = 1/object_distance + 1/lens_distance
and the mirror was likely offset by half of the difference in lens distances. For example, if you were using a 50mm lens, and the pairs are (infinity:20m) and (3m:2.5m), the difference in lens distances is about 0.15mm, and the mirror was probably off by ~ 0.075mm, but the sign of offset depends on the previous question.

It's unclear to me if that tells what you need to know, because it's how much the previous mounted-with-sticky-tape position was off by.

 

[Ziotoo4]

Thank you reddesert, this is fantastic and exactly what I was looking for.
Apologies for my unclear explanation - english is not my first language. To clarify:

• when I point the camera to an object at infinity (e.g. the moon), I cannot extend the lens focus enough to have the focusing screen prism show the object as in focus.
• when I point the camera to an object 20m away (measured with laser range finder), the focusing screen prism and the autofocus show the object in focus when the lens is set at infinity.
• when I point the camera to an object 2.5m away (measured with laser range finder), the focusing screen prism and the autofocus show the object in focus when the lens is set at 3m

So in other words, if I follow the indication of the focusing prism, I am focusing behind the actual location of the object.
For instance, in this shot I focused on the target, which was 2.5M away. The target was in focus in the viewfinder. But the lens distance scale showed 3 meter and sure enough the actual focus point is 50cm behind the target.

So now, if I understand the formula correctly:

1/f = 1/object_distance + 1/lens_distance

for the pair (2.5m object distance, 3m lens scale distance) we get:

2.5m -> 1/50mm = 1/2500mm + 1/X -> 49/2500mm = 1/X -> X = 51.0204 mm
3m -> 1/50mm = 1/3000mm + 1/X -> 59/3000mm = 1/X -> X = 50.8474 mm

Therefore the delta is 51.0204-50.8474 = 0.173mm, which we have to divide by two because of the mirror bounce, which means the sticky tape additional thickness should be 0.865mm

Would this sound about right?

 

[r_a_feldman]

0.173/2 = 0.0865, not 0.865

0.0865mm is about 0.0034 inches

Something like Scotch 665 might work, or might be a bit too thick.

 

[Dan Daniel]

0.173/2 = 0.0865, not 0.865

0.0865mm is about 0.0034 inches

Something like Scotch 665 might work, or might be a bit too thick.

You can also use a thin 'plate' and apply tape to both side to get the right final thickness. So a thin transfer tape at about .010", 0.25mm, twice, with a 0.25 shim stock will get you close. Etc. Vary shim layer to get the proper thickness.

To measure assemblies with tapes it is best to make a full assembly, roll down, and measure the full assembly. Many tapes apply thinner or thicker than the 'spec' thickness.

 

[reddesert]

The lens is actually focused at 3m while the focusing screen indicated 2.5 m. This means the film was closer to the lens than the focusing screen was. The focusing screen not reaching infinity when the lens is at infinity says the same thing (just as a check).

Thus, you need ~ 0.086 mm of extra thickness beyond whatever was on there when you took the photo (was that the sticky tape from the previous repair?) As r_a_feldman noted, you dropped a decimal place at the final step.

I think the backing paper from 120 film is about 0.1mm thickness, to give an idea of how thick/thin this is. You aren't going to be able to get it accurate at the ~0.005 mm level, so don't worry too much about the third decimal place.

(Roughly speaking, for a circle of confusion of 0.03mm and photographing at f/2, one would like to get the lens-to-screen distance accurate to better than 0.06mm, which means the mirror depth to better than 0.03mm due to the reflection. That's for acceptable sharpness since it used all of the circle of confusion, and one would like to get it maybe a factor of 2-3 better for critical sharpness. Obviously, if you stop down a little, this gets easier. This is one reason I find the "I need to take all my photographs at f/1.4" trend a bit questionable.)

 

[r_a_feldman]

0.173/2 = 0.0865, not 0.865

0.0865mm is about 0.0034 inches

Something like Scotch 665 might work, or might be a bit too thick.

I just got home and measured an old roll of Scotch 665 that I have. It mics out to about 0.0032"-0.0034".

 

[Ziotoo4]

Hi all,
after applying 0.09 thick tape, the problem is significantly reduced and almost fixed, but alas, still ever so slightly off at infinity.
I have ordered some industrial grate .1 thick tape since I suspect the 0.09 one was 'roughly' 0.09.

Will update once I receive and applied the tape.

 

[brbo]

I'm 99% sure that camera has ways for fine adjustments of the mirror position. Solving this with finding a tape with just exact thickness is probably not the way to do it.

 

[monopix]

Hi all,
after applying 0.09 thick tape, the problem is significantly reduced and almost fixed, but alas, still ever so slightly off at infinity.
I have ordered some industrial grate .1 thick tape since I suspect the 0.09 one was 'roughly' 0.09.

Will update once I receive and applied the tape.

I've refitted a couple of AX mirrors and several more RX mirrors and have always used 0.1mm tape. It usually works. But I have seen variation in mirror thicknesses in some models (not the AX or RX) so no guarantees.

 

[Ziotoo4]

I'm 99% sure that camera has ways for fine adjustments of the mirror position. Solving this with finding a tape with just exact thickness is probably not the way to do it.

Leaving this here since this would be my conclusion also, but it's incorrect brbo

The official repair manual says that focus adjustment must be done by opening the camera, shimming the washers under the viewfinder, and then corresponding adjustment via software for the AF CMOS.

The fine adjustment for mirror position is only for very small adjustments (mostly due to variation in mirror thickness)

The reason for this is, there is a secondary mirror that reflects down to the AF module (see items 2 and 7 below) so you can't adjust the position of the mirror or the AF slit wouldn't align.

The repair manual says, if the mirror ever unsticks from the carrier, call contax and get a new mirror assembly, but that's not really an option here.

Thanks all for the advice, will post here when the new tape comes and I attempt the repair again!

 

[monopix]

I'm 99% sure that camera has ways for fine adjustments of the mirror position. Solving this with finding a tape with just exact thickness is probably not the way to do it.

I'm pretty sure there isn't but, if there is, you don't want to mess with it because the autofocus uses a secondary mirror attached to the back of the main mirror and if you adjust the position of the main mirror you'll fcuk up the autofocus.

Always amazes me how people are so ready to give advice about things they know nothing about.

 

[brbo]

Ok, so the Contax repair manual that has a section (as I predicted) about fine adjustments was written by someone that knows nothing about that camera.

Good to know!

 

[Don_ih]

The fine adjustment for mirror position is only for very small adjustments (mostly due to variation in mirror thickness)

That should also be enough to adjust for variation in adhesive thickness, shouldn't it?

 

[Ziotoo4]

That should also be enough to adjust for variation in adhesive thickness, shouldn't it?

Manual says, you can adjust at most half the length of the infinity symbol on the lens.

As per initial discussion however, the distance is off by roughly half a meter on the 3m scale.

If you calculate using the formula and looking at the distance scale on the planar, the screw should be able to adjust for roughly .01 mm in thickness variation.

So I'll first use the 0.1 mm adhesive as recommended by monopix, and then do the fine adjustment with the screw indeed. Finally, I will adjust the AF

 

[brbo]

after applying 0.09 thick tape, the problem is significantly reduced and almost fixed, but alas, still ever so slightly off at infinity.

As per initial discussion however, the distance is off by roughly half a meter on the 3m scale.

Which one is it then?

Anyway, of course, you should get mirror and adhesive thickness as close to the original. Failing that (and you will fail, even if so slightly), you should use fine adjustment provisions as implemented by the manufacturer for perfect adjustment. Unless, you are fine with just... fine.

 

[Ziotoo4]

Having received and installed the new tape, the focus is now less than a quarter of the infinity sign turn off.
So I wanted to proceed to adjust the mirror stop, but I realized I do not know how to turn this screw, since I can't properly reach it with my JIS screwdrivers.

The screw is pretty tight, and it is dangerously close to both the shutter and the mirror. There is no way to hold the shutter open aside from bulb mode, so accessing from the back is not an option or I risk damaging the shutter; anyhow the screw is equally recessed from the back side as it is from the front side, so I still have the problem of how to turn it with my (straight) screwdriver.

How to turn this screw at an angle? What tool should I look for?
Thank you so much for your advice.

 

[Don_ih]

Is there supposed to be a rubber O ring on this screw?

I may have it confused with some other kind of camera.

If you can get a long-ish philips driver that's the right size, you can heat it up with a propane torch and bend it to a right angle. Heat it again and plunge it it water, or the tip will lose its hardness (which it may do, anyway).

 

[reddesert]

I don't think it's literally a philips head screw - the slots go all the way to the edge.

I would take a long, narrow flat head screwdriver and try to reach it at an angle. Because it is an adjusting screw and not a fastening screw, hopefully it won't take much torque to turn. There are L-shaped (offset) screwdrivers for awkward reaches like this, but they're usually much larger. Another option is, if you can find a screwdriver bit small enough to fit that screw, hold the bit in a wrench to turn it.

 

[Dan Daniel]

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

But really, a long small screwdriver as reddesert says is your quickest way forward.

 

[Ziotoo4]

An update (and I think if there's no suggestions this might be the last update, since I'm completely lost on how to finish this repair).

Using both a long thin screwdriver as well as an offset (L-shaped) screwdriver I reached the screw.
The thin screwdriver was a JIS + screwdriver and for the L-shaped I jerry-rigged a JIS + point and an L-shaped wrench.

Neither of these was able to turn the screw, not even one bit. Some of the screw paint chipped off but the screw itself won't budge. I applied acetone with a q-tip and waited and tried again, but nope, the screw will not budge.
The manual says to apply a mysterious 'cemedine 551' after adjustment, and I assumed this was some sort of epoxy, but I cannot see it anywhere on the screw and acetone 99.99% didn't seem to help.

I also assumed that the adjustment screw would need minimal torque to move, but that's not the case. I am about to strip the screw's head, and it still won't turn.

What am I missing here?

 

[europanorama]

i have a most probably never used AX-museum? it has clearly backfocus. will now compare groundglass. using OM-screen tilted is best oleson said. with viewfinder. also have loupe and split. have extensive experience checking focus on Kiev 60 and P67-lenses. and reaching infinity with digital. go to pentaxforums for that.
for Contax AX go to FB Yashica & Contax users. i gave links to german specialist who has 3 technicians.

CONTAX

Contax Kameras Zeiss Objektive 1974-2005

contax-cameras.reconact.com

check service. i already asked if they can adjust backfocus.
2. dont forget. Focusshift when stopped down. And: Is your lens reaching infinty? maybe it need collimation.
Infinity: 8-10km lensmakers confirmed. seen tiny difference and yes have novoflex-lenses which focus beyond and can be adjusted. among tested lenses where P67 300/200. also 300 green-star- yes adapter-reliability also relevant. novoflex has special adapter which is shorter to reach infinity. and i have an adjustable register adapter created. discussion at mflenses and pentaxforums. more when i have more time.
will compare with my new zeiss 300/4 MM also with Leitax EF-mount, made shorter to reach infinity with digital.
analoge MF lenses on digital do not reach infinity experts confirmed.
struggeled a lot testing AF on AX. have professional Canon EOS SRL for comparison using Dandelion Lushnikov chip. AF-confirm with EOS seems a lot better.
only comparision of viewfinder with Groundglass(never use a plastic selfmade one, if ever real glass by photofinder, yes much more coarse than OM or Ni-screens.) will tell if i have mirrorslipping.

 

[europanorama]

heat is needed to open that screw. let the germans do that. they have a forum to get an offer. will let you know how that is working in a few days.