Canon T90: Service and repair using practical examples

Interestingly, the LED lights up at just 1 volt of forward voltage. I'm familiar with values of 2 volts and higher, but I'm no expert.

The structure of the LED is also interesting.

I wonder if this isn't a light bulb with a filament?

Interaction between mirror, aperture control and FDn lenses

To better understand the interaction between mirror, aperture control and lens, I take a look at the removed mirror box (not complete) of a T90.




Front view, the mirror is cocked and ready to fire.




Aperture control, rewind site.




Mirror mechanism, release side.




View from above.




View from below.




I use the probe to pull the armature away from the release electromagnet.

This corresponds to the current flow through the coil on the magnet when it is triggered, which releases the armature.




The mirror has triggered.




I tension the mirror.




The mirror is folded down and ready to be released again.




Now I attach a FDn 50/1.8 to the mirror box.

The aperture ring is in the automatic position (A).

The mirror is still tensioned.




Front view, with the mirror folded down.

The subject can be viewed through the viewfinder (no longer present here).




View of the aperture control.




Using the probe, I release the aperture via a lever (white arrow).

The wiper moves down over the flush plate (magenta arrow).

The aperture control locking pawl is not engaged yet (blue arrow).




Now I release the mirror.

The locking pawl engages.




The aperture is closed to the minimum value (f = 22).

Now I set aperture f = 4 manually.




I release the aperture.

The wiper moves a little over the flush plate.

The locking pawl has not engaged.




Now I trigger the mirror, the locking pawl engages.




The lens is stopped down to f = 4.


My understanding of the sequence in automatic mode (A) is

• that, when the T90 is triggered, the aperture first closes.
• To do this, the aperture electromagnet is triggered electronically, releases a lever which clears the wiper moving down the flush plate.
• I triggered this lever manually with the probe.
• If the aperture is larger, the wiper moves less downwards.
• If the aperture is smaller, the wiper moves more downwards.
• The locking pawl secures the wiper in the respective position as soon as the mirror is released via the release magnet.
• In short, the aperture magnet starts the closing of the aperture and the release magnet stops it at the right time, depending on the aperture value.
• The gold-plated contact track on the flush plate, together with the movable wiper, reports the respective aperture position to the control electronics of the T90.
• The whole process is synchronized and takes place in milliseconds.

The manual mode with aperture pre-selection on the lens should only differ in that the aperture control clutch on the lens controls the position of the wiper on the flush plate mechanically.

Here are the FDn coupling elements in the mirror box and on the lens:





Yellow: stop down lever in the mirror box and corresponding stop down clutch on the FDn lens.

Turquoise: aperture control lever in the mirror box and corresponding aperture control clutch on the FDn lens.

I attached the bayonet ring of a T90 to the lens to simulate the position of the aperture when the lens is attached.


The stop-down lever in the mirror box (yellow) always moves fully toward the release side when the shutter is triggered.

The aperture control lever in the mirror box (tourqoise) determines the aperture setting and has two modes:

1. In automatic mode (A on the aperture ring of the lens), its movement is electronically controlled by the T90. It moves less for large apertures and further down for small apertures.
2. In manual mode (aperture selected on the lens), the aperture control clutch on the lens moves to the appropriate position, thus setting the aperture.

---

That all is my assumption, unfortunately I have not found any description of these sequences and correlations in my technical documents.

Aside from general interest, these insights are important for troubleshooting a problem on the T90 featured in this thread.

The aperture only closes to f/8 in both automatic and manual mode.

When I remove the mirror box, I now know that I have to inspect the aperture control lever, its mechanics, and its controls.

The position of the two electromagnets on the mirror box of the T90 (rewind side):

• release magnet (turquoise)
• aperture magnet (magenta, partly hidden)

More on this soon in my repair report.


+++

All information provided without guarantee and use at your own risk.

AVO* adjustment: Setting the maximum lens aperture on the T90

In Av (aperture priority) mode, the lens aperture is set electronically using the control wheel. The lens aperture ring is set to „A“ for automatic.




In order to calculate the exposure, the T90 needs to know what the maximum aperture is.

In the FDn system, this is transmitted from the lens to the camera via the full aperture signal pin on the lens bayonet.

As a counterpart the maximum aperture control pin is located in the mirror box of the T90 (green arrow).






This pin is pressed in by the signal pin on the lens more or less depending on the light intensity of the lens.

The different high of the signal pin stands for the corresponding maximum aperture value.


Flush plate

On the flush plate on top of the mirror box there are contact tracks (switches) and wipers for transmitting aperture settings and maximum lens aperture to the electronic control system of the T90.




The flush plate with contact brush on contact track, which transmits aperture values to the electronic camera control with its respective position.




Flush plate removed.




AVO contacts (switches) on a second board located behind the flush plate with wipers for transmitting maximum lens aperture to the camera.


The AVO system can be adjusted

using a screw in the maximum aperture control pin.

The Canon service manual provides height information for the pin, which is measured with a depth gauge to an accuracy of 1/100 mm. Alternatively, „known good“ 1.8 and a 2.0 FDn lenses can be used.

The screw is then adjusted until the correct maximum aperture (light intensity) of the lens is displayed on the LCD of the T90.

On my T90 featured in this thread, the small apertures are not closed far enough.

When checking the AVO setting, I saw that in Av mode my FDn 50/1.8 can only be set up to f = 2.0.

I corrected this by adjusting the maximum aperture control pin.




To do this, a small drop of acetone has to be applied to the screw to loosen a seal.




After a few minutes, the screw can be turned using a 0.9 mm flathead screwdriver. Be careful, it is made of soft brass.




After the correction, the maximum apertures on my FDn 50/1.8 and 35/2.0 are displayed correctly.

This has also improved the formation of the small apertures in Av and Tv modes. When set manually, the openings are still too large.


To do

Larry Lyells in the SPT journal:

For erratic exposures, you can
check the light transmission in AV
mode. If the diaphragm opening is
inconsistent, the problem is
probably a dirty aperture magnet or
flush plate. If the diaphragm
opening is consistent, check the
shutter speeds in TV mode.
Inconsistent shutter speeds may indi-
cate dirty shutter magnets.

Click to expand...

I will therefore check the AVO switches with a multimeter and clean them, as well as the switches on the flush plate, the associated wipers and the aperture magnet.

I will also look at the function of the aperture control mechanism; I have gained experience with this in the last few days.








The binary values of the AVO switches can be measured at their cable connections (red, brown, grey, yellow) on the flexible circuit board. For this the mirror box does not have to be removed.

There should only be two clearly separated voltage ranges for high and low voltage to measure.



A table with the corresponding binary values for the connections can be found in the Canon Service Manual.


Additionally, the T90 switches to EEE/Help when setting smaller apertures in AV. There may be a connection, and/or the shutter may be stuck.

I'll check this when I disassemble the camera.

___

*AVO stands for „Maximum aperture value correction sensor“, see




+++

All information provided without guarantee and use at your own risk.

Cleaning aperture magnet Mg1

In order to clean the aperture magnet, the mirror box must be removed.




I took a closer look at this on a mirror box that lacks the upper structure with the prism. But the structure around Mg1 is complete.




To distance the armature from its two contact surfaces, the mirror is released. This is done via the release magnet Mg2 by pulling back its armature.

The aperture magnet Mg1 is an electromagnet that doesn't attract its armature without current, as is the case here.

In contrast, the release magnet Mg2 is a permanent magnet that releases its armature when current passes through its coil.




The mirror has triggered and flipped up.




I turn the mirror box upside down.




Rear, right corner, top

• Mg1: green arrow
• Mg2: magenta arrow

Mg1 can be cleaned from the outside without dismantling.

The arrows point to its two contact surfaces under the armature.




To get between contact surfaces and armature I bend a cleaning swab.




For cleaning I use benzine.






The area can be easily reached and cleaned with the swab.

With a bit of luck, this might solve my problem with the small apertures.

We'll see


+++

All information provided without guarantee and use at your own risk

Removing the auto-diaphragm unit

This allows, among other things, the aperture magnet Mg1 to be reached in order to have full access for cleaning or replacement.

Larry Lyells:

2. To remove the auto-diaphragm unit, Fig. 11, remove the cemented indicator plate at the bottom front of the mirror box (the plate with the red dot next to the diaphragm-closing lever). Take out the 3 screws holding the auto-diaphragm unit (1 long screw, 2 countersunk screws). You can then lift off the auto-diaphragm unit with the release magnet and reach the SP flex and the aperture magnet.

Click to expand...

3 screws, one under the movable fitting.




Removed






Mg1, 2 screws




The electromagnet and …




… its armature.

Removing the AE unit

This allows the AE unit to be removed for cleaning and lubrication or replacement. The aperture magnet Mg1 is also accessible.

Larry Lyells;

To remove the AE unit, Fig. 12, first take out the AE-brush guard (screw at front, under wires). Then remove the 2 screws holding the AE unit.

Click to expand...

This process worked for me:



Release the mirror by pulling out the armature of release magnet Mg2.

The mirror will flip upward.




Position of the two retaining screws.




Loosen the lower retaining screw.






Removing the AE-brush guard, 1 screw.




In this position the aperture control lever can be threaded out of the mirror box.




To do this, release the locking pawl (lower tip of the double arrow) by pulling and holding the corresponding lever (green arrow).

The aperture brush (upper tip) can now move to the appropriate position via the large gear; it is coupled to the aperture control lever.

Release the lever to fix the aperture brush.

Then loosen the upper retaining screw (magenta arrow, here already done).




Done

Now I should be able to search for the problem with the small apertures without any hindrance; spare parts are available if needed.

Stay tuned!


+++

All information provided without guarantee and use at your own risk.

AE unit removed, cleaned, oiled and lubricated, checked







In order to thoroughly service the AE unit, I removed it as described previously in the thread.




This also made aperture magnet Mg1 accessible, which I dismantled and cleaned with benzine.






Cleaning all contacts and wipers with electronic cleaner.




I oiled the gear axles, removed old grease with benzine, and re-lubricated sparingly.




After that, the AE mechanism moved smoothly.




Everything worked with the lens attached.

I couldn't test whether this had solved the problem with the small apertures until after I had reassembled the T90.

I suspected it was Mg1, which, as an electromagnet, requires electricity to function.


The first not sticky shutter I have encountered



I couldn't believe it. The damper in this shutter is intact. It hasn't disintegrated into a black, sticky mess.

Until now, I've only seen sticky shutters on the T90.




The two plastic worm screws for tensioning the shutter curtains are still intact. However, the top screw is bending under the spring pressure and will likely break. Therefore, I cannot adjust this shutter.






The table stand allows you to work comfortably.


And once again EEE/HELP

After assembling the camera, the already known error occurred.

When switched on, the charge motor jerks and the camera displays EEE/HELP.

I was unable to find the error.

I worked through the tests following the C & C troubleshooting instructions.

There is obviously an open spot on the path from IC3 (pin 19) via front flex connector and switch SW13 on the mirror box to ground. The path should have a low resistance with SW13 closed. Anyway I was able to successfully test SW13 for continuity but not from IC3 to the front flex connector.

SW13. Mirror switch on mirror box,
… (direct connection to SP flex
- connects to IC3 pin 19 through
front flex connector). Normally
closed, connecting pin 19 to ground.
Opens when the mirror is up, telling
IC3 to release the shutter.

Click to expand...

Strangely, there is no full continuity with spare parts T90 either.

Desoldering and checking two double diodes also revealed no errors.

Adjusting the two retaining screws on the front flex connector also did not bring any success.

I don't understand it




Finally bridging IC3 to the front connector with enamelled copper wire did not help either.

---

So this T90 is also going into the spare parts inventory.

Of course, that's frustrating, because I couldn't determine whether I'd solved the aperture problem. And a lot of work remained unsuccessful, and I don't know what the cause is.

This is the third time that EEE/HELP has killed my project after assembly

But I was able to learn something new and achieve some success, see thread


+++

All information provided without guarantee and use at your own risk.

In a tutorial, everything should work out. That's not the case here. Therefore, I'll update the thread title.

Here, the gold-plated contacts are partially worn.

This is the AVO switch, which transmits the maximum aperture of the attached lens to the camera control system.

To do this, the lens presses the maximum aperture control pin in the mirror box in a varying degree. This mechanical movement is transmitted to the AVO switch, which moves along the pattern of contact paths and generates a 4-bit code of 0 (non-conductive) and 1 (conductive).

The switch thus acts as an analog-to-digital converter.

If this switch isn't working properly, the camera will receive an incorrect maximum aperture reading, which could possibly explain why the small apertures aren't being formed.

But this is hypothetical,

EEE/HELP: Challenge in the area of the front flex connector of the T90

Since there have been repeated problems with this variant of the EEE/HELP error after disassembling and reassembling a T90, I will focus on it. Otherwise, all other work on the T90 is pointless if it no longer works.

According to technical documentation and my own experience, EEE/HELP and bc can have various causes.

The T90 contains a variety of different switches that are activated during the execution of a shutter sequence, for example.

A switch is anything that either conducts current or not, which corresponds to the digital camera control with 1 HIGH and 0 LOW, such as the mirror switch, SW13:

SW13. Mirror switch on mirror box,
Fig. 13 (direct connection to SP flex
- connects to IC3 pin 19 through
front flex connector). Normally
closed, connecting pin 19 to ground.
Opens when the mirror is
up, telling
IC3 to release the shutter.

Click to expand...

Depending on the symptoms, the SPT Journal and the C & C Troubleshooting Guide offer different troubleshooting routines for these error codes.




In my case, it seems to be a problem with the assembly of the three circuit boards in the front flex connector.




Here, three circuit boards are sandwiched together, with narrow elastic connector strips in between.






They are held together by a solid metal pressure plate that is slightly curved and secured with two screws. These screws must be tightened to the correct ratio to ensure the contact is not interrupted or damaged.


What's puzzling me is measuring a line from IC3, the central CPU, through the front flex connector and the aforementioned switch SW13 to ground.

There should be continuity (0 ohms, low resistance) here when the mirror is tensioned and the camera is ready to fire.

However, that's not the case with any of my spare T90s.

I can measure continuity from one board section to ground via SW13, but not continuity from IC3 through the other board section.

This means that the problem is not with the contact in the front flex connector, but rather in front of it.

This is what I will focus on to solve this problem, it is the key to all further repair attempts.




Here's a block diagram from the C & C guide showing this circuit section (red mark):

• From IC3 at pin 19, the path goes via switch SW13 to ground.
• When SW13 is closed, there should be continuity here (~0 ohms).
• SW13 then sets pin 19 to LOW.

I'm not sure what the small rectangles in between mean.

Can anyone shed some light on this?

Andreas Thaler said:

The structure of the LED is also interesting.

I wonder if this isn't a light bulb with a filament?

Click to expand...

It's clearly not an LED which you could have determined by verifying whether it's a diode (which you probably would have told you it isn't) and by the fact that white LEDs didn't exist when this camera was made.

Andreas Thaler said:

I can measure continuity from one board section to ground via SW13, but not continuity from IC3 through the other board section. This means the line must be open or high-resistance somewhere in between.

Click to expand...

It's unclear to me between which points you're measuring exactly. If you could make a diagram that indicates this, it would be easier to comment.

Andreas Thaler said:

I'm not sure what the small rectangles in between mean.

Click to expand...

I expect those represent the interconnects between different physical parts (boards, flex PCBs etc).

koraks said:

It's clearly not an LED which you could have determined by verifying whether it's a diode (which you probably would have told you it isn't) and by the fact that white LEDs didn't exist when this camera was made.

Click to expand...

In the technical documentation it is consistently referred to as LED2, I will check if the current flows in both directions.

koraks said:

It's unclear to me between which points you're measuring exactly. If you could make a diagram that indicates this, it would be easier to comment.

Click to expand...

I'll do it when I have the T90 at the table, along with a photo.

koraks said:

I expect those represent the interconnects between different physical parts (boards, flex PCBs etc).

Click to expand...

Thanks, that's my view too. However, I don't understand why SW13 is in between here. The top-down order is IC3 - front flex connector - SW13 - ground.

Andreas Thaler said:

In the technical documentation it is consistently referred to as LED2, I will check if the current flows in both directions.

Click to expand...

Visually you can already tell it's a filament of some sort. Filament LEDs only came around about a decade ago or so.
Also, the supply voltage you found obviously points towards a filament light source as 6V is/was a common supply voltage for them (remember those old bicycle bulbs?)

Andreas Thaler said:

Thanks, that's my view too. However, I don't understand why SW13 is in between here. The top-down order is IC3 - front flex connector - SW13 - ground.

Click to expand...

Basically SW13 is unconnected on whatever assembly it's part of, and both of its connections are made on the board that houses IC3. From a viewpoint of modularity/commonality this makes good sense to me as another implementation around IC3 could choose to e.g. pull up SW13 instead of pulling it down.

Also, is it possible your thread went off track a little? The title suggests it was going to be a tutorial, but it looks like you've gone into regular "feel my way around and see if I can fix something" mode.

koraks said:

Also, is it possible your thread went off track a little? The title suggests it was going to be a tutorial, but it looks like you've gone into regular "feel my way around and see if I can fix something" mode.

Click to expand...

In a tutorial, everything should work out. That's not the case here. Therefore, I'll update the thread title.

Click to expand...

See above.

I write here because I enjoy it, but feedback like this doesn't contribute to that or motivate me. I don't understand the point of it either, it makes me tired.

Andreas Thaler said:

See above.

I write here because I enjoy it, but feedback like this doesn't contribute to that or motivate me. I don't understand the point of it either, it makes me tired.

Click to expand...

It's pioneering work, keeping the hobby alive and all that, provided for free. No, not free, provided at great expense in your own time and money. Unfortunately you got a little clunky in your wording so we're going to have to send you to the courts to stand for your crimes.

I've thought this through carefully.

Yes, it was planned as a tutorial, but with every T90 there are always new problems that need to be addressed. That's how it is in practice. Hence the deviation and my correction of the thread title.

I think I've said enough about the T90; see also my current post.

That's the only T90 project documentation on the web, by the way. I'm glad about that, but I don't take it too seriously. Without Larry Lyells/SPT, C & C and the service manual, I wouldn't have been able to write anything.

Have a nice weekend, I'll be continuing with the Canon A-1 in the summer. I'll try to fit everything into one thread, so I agree with @koraks on that

I'm currently looking for something on Japanese websites about T90 repairs.

Here's a blog post from a Japanese colleague who guides through disassembling a T90 to remove the sticky damper in the shutter. The battery compartment of the camera is contaminated with battery electrolyte.

He also uses an interesting electrolysis method to remove the copper plating from housing parts and replaces a broken plastic adjusting screw in the shutter with a metal one.

I was able to translate the pages sufficiently using the translation function in my iPhone's Safari browser to follow along.