Canon T90: Solution found for the stubborn front flex connector, service and repairs on the T90 are no longer a gamble

[Andreas Thaler]

It's happened to me several times now that I've serviced a T90, only to find the work was in vain because of unsolvable contact problems in the front flex connector.

The front flex connector is located on the front release side, where the red self-timer LED is located. Three flexible circuit boards are connected here.

Elastic connectors with small gold-colored inclusions rest on the gold-plated contact rows.

The sandwich is hold together by a metal pressure plate with two screws.


To remove the mirror box and shutter, the front flex connector must be opened.

After assembly, the pressure on the retaining plate is controlled by the two screws.

This worked on some T90s, but not on others. If there are contact problems, the camera goes crazy and displays error messages.

This made servicing and repairing the T90 a gamble.


Two ways to bypass the front flex connector

So I considered how I could remove the mirror box and shutter without opening the front flex connector.

I found two options:

1. Unsolder all the cable connections from the mirror box to the circuit boards above it and detach them from the mirror box. Then the mirror box can be lifted out of the housing.
2. Or leave the mirror box connected to the circuit boards and lift it out of the housing only far enough to access its mechanism and the shutter underneath.

I opted for the second solution because I had already done some preliminary work on it, and the mirror box's wiring is very dense. I would also have to disconnect the interconnected circuit boards. Too much effort and risk.

The disassembly of the T90 follows Larry Lyell's instructions in the SPT Journal. Only the step involving opening the front flex connector is omitted, and I've modified the procedure.

---

The status of the repair candidate is unclear because the battery compartment and battery contacts are contaminated with battery electrolyte and the camera is probably therefore not showing any signs of life.

Battery compartment removed.

Battery electrolyte on the tripod attachment.

But the battery cables are OK, so I don't need to replace them.

Connected to my lab power supply, the T90 comes to life. All functions are there, as far as I can tell.

This would allow me to reassemble the camera and, after cleaning the battery electrolyte, put it back in the closet.

But the sticky damper in the shutter is probably still there, which needs to be removed, otherwise it will clog the shutter blades and render the T90 unusable. As is probably the case with most T90s these days.

So I remove the mirror box following the new procedure to get to the shutter, which I also take out to disassemble and clean it.

Desoldering the cables from the circuit board to the top cover.

Removing the trigger unit.

Remove the spring on the stop down mechanism.

Desolder and loosen the circuit board connections on the rewind side above.

 

[Andreas Thaler]

Desolder the red and grey LED cables …

… as well as the red and black cables for the backlight of the main LCD …

… and the black, light blue and yellow cables below the main LCD.

Remove the eyepiece.

Unscrew the ground connection for the shutter.

Desolder the black cable …

… as well as the shutter connector.

Now, after loosening its four retaining screws, the mirror box can be carefully lifted out of the housing and folded toward the release side.

Be careful not to tear the flexible main board with which the mirror box is tightly connected.

The mirror box is fixed with adhesive tape.

Pull the shutter connector out from under the main board.

Loosen the three retaining screws for the shutter. For the screw on the release side, top, you do this before lifting out the mirror box to gain access.

I just learned that

Some T90s have an elastic fastener that additionally secures the shutter to the housing. This can be cut with a sharp knife.

Pry the shutter out carefully with a flat-head screwdriver, as the knife does not reach everything.

Now it takes patience and caution to remove the shutter.

The spot where the sticky damper sticks out is handy for this. I hook the angled probe into this.

That is safe, the probe engages directly into the stable shutter frame, there is only a part of the damper, which is removed anyway,

The shutter is pulled out of the housing.

Desolder the orange cable that supplies power to the shutter onto the motor flex at the bottom of the camera.

 

[Andreas Thaler]

Done, the shutter is out.

The orange cable is desoldered from the shutter here. It's better to leave it in place and unsolder it from the circuit board; that's easier to access.

Disassemble and clean the shutter.

Remains of the damper for the two shutter curtains that has become soft and sticky.

Clean with acetone.

One of the two plastic screws for adjusting the spring tension for the curtains has broken. This is a common occurrence with the T90 these days.

As a result, the shutter can no longer be adjusted. However, the greenish screw seal should hold it in the factory-set position.

If the screw breaks completely, the spring tension might drop and the shutter will be defective.

Some oil for the two curtains bearings, one visible here.

The shutter is cleaned, oiled and reassembled.

Cleaning the magnets with benzine.

I use a multimeter to check the continuity of the two gold-plated switches at the curtains.

Everything is fine, but I still clean the switches with electronic cleaner.

I test the two magnets with 3 volts from the lab power supply to see if they trigger.

They do.

To protect the delicate curtains when installing the shutter, I manually release the first curtain (opener). The curtains are now folded in.

To get a little more clearance when inserting the shutter into the housing, I unscrew the rewind motor.

 

[Andreas Thaler]

Now comes the most difficult part, which requires perseverance, patience, and caution. The shutter must be inserted into the housing and the charge lever must engage with its counterpart.

It's a game of patience, but after many attempts (I practiced on a spare T90 beforehand), it works.

This picture is from a different T90; I forgot to take a photo.

I connect the shutter to my lab power supply to check if the charge motor is correctly tensioning the curtains. I also check the release magnet.

Larry Lyells describes all the necessary steps in the SPT Journal (see link above).

These circuit board connections aren't easy to solder. I use a multimeter to check for continuity.

The command wheel contacts and the switches in the trigger unit are treated with electronics cleaner. This also eliminates the usual command wheel malfunctions.

Done.

The T90 is installed and everything works.

The front flex connector was tricked; it didn't ruin my work this time

 

[Andreas Thaler]

Conclusion

• I still don't fully understand the function of the elastic connectors in the front flex connector. They undoubtedly provide contact; they're supposed to conduct, but they don't short-circuit the row of contacts.
• I still have to figure out why contact problems occur after opening the front flex connector. Perhaps the elastic connectors have changed their properties after decades of service, or Canon had a special method for adjusting them.
• Since the symptoms are reminiscent of problems with the ground connection (erratic camera functions, unstable, not or only with difficulty reproducible), it could also be a problem with the soldering of the circuit boards on the connector. Perhaps a certain retaining pin also has a ground function and I repeatedly soldered it incorrectly.
• In any case, this alternative method for removing the mirror box and shutter works.
• This means that service and repair work on the T90 are no longer a gamble

+++

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

 

[Andreas Thaler]

Excellent shutter performance

Both shutter curtains move at practically the same speed, resulting in minimal exposure deviation in the image.

1/4000 second

1/60 s


The target curtain travel time is 2.7 milliseconds, which puts this T90 somewhat on the slow side.

 

[Andreas Thaler]

View attachment 400310

One of the two plastic screws for adjusting the spring tension for the curtains has broken. This is a common occurrence with the T90 these days.

As a result, the shutter can no longer be adjusted. However, the greenish screw seal should hold it in the factory-set position.

If the screw breaks completely, the spring tension might drop and the shutter will be defective

I could have permanently fixed the partially broken screw with Loctite. That would have eliminated the risk of a complete break.

But this way, I can also monitor how things develop.

I have a spare shutter from another T90.

 

[Andreas Thaler]

If the mirror box is removed, you should take the opportunity to have it serviced, see

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?

www.photrio.com

 

[forest bagger]

A fine work, Andreas!
And to the problems with the elastic connectors in the front flex connector: It may be of interest to know how they work and how to adjust them after first removal, but what the heck - your work around is good.

 

[Andreas Thaler]

A fine work, Andreas!
And to the problems with the elastic connectors in the front flex connector: It may be of interest to know how they work and how to adjust them after first removal, but what the heck - your work around is good.

Thanks, Michael

 

[Andreas Thaler]

And to the problems with the elastic connectors in the front flex connector: It may be of interest to know how they work and how to adjust them after first removal,

https://phsc.ca/repair/Canon/canon_camera_t90.pdf

The C & C Guide shows the structure of the front flex connector with the elastic connectors (pads).

This is nothing new so far, but perhaps it will provide an explanation of how the strips work.

Pressure plate with two sealed screws mounted.

Loosened.

Here you can see the contact rows of the front flex connector with spaces in between as slight indentations. The elastic connectors are inserted.

If the strips are positioned at a right angle over the contact rows and pressed down with the pressure plate, they should become wider and thinner.

This also presses the gold-colored metal particles interspersed in the strips together, making the strips conductive.

They then electrically connect the three contacts lying one above the other.

The spaces ensure that the pressure is less there, which keeps the metal particles apart and prevents any electrical connection. Otherwise, the contact rows would be short-circuited, which is not the point.

Everything can only work if

1. the pressure is sufficient to make the strips conductive and
2. all contact rows and the pressure plate lie flat on top of each other to avoid contact problems. And that seems to be the weak point.

I'll continue to investigate this with my spare T90s and try to verify this hypothesis and conduct resistance measurements with the elastic connectors.

For practical purposes, I'll stick with the alternative method shown, removing the mirror box and shutter, as it avoids contact problems by opening and thus disturbing the working front flex connector.

Nevertheless, it would be interesting to know how Canon adjusted the front flex connector during production. There are only two screws on the mounting plate for this.

 

[Andreas Thaler]

Investigating the elastic connectors: assumption confirmed

I'm measuring the resistance of an elastic connector strip from one end to the other with a multimeter.

To approximate the conditions in the front flex connector, I use wide measuring tips.

The strip has a high resistance at this distance, conducting only minimal current.

My multimeter shows up to 2 megaohms, which is exceeded.

Here's a resistance measurement through the strip.

This corresponds to the position in the front flex connector, where the strips are located between the rows of contacts.

With light pressure of the test probes against each other, the strips have a low resistance.

Here, the smallest setting is 200 ohms on the multimeter.

This confirms the assumption made above:

If the strips are positioned at a right angle over the contact rows and pressed down with the pressure plate, they should become wider and thinner.

This also presses the gold-colored metal particles interspersed in the strips together, making the strips conductive.

They then electrically connect the three contacts lying one above the other.

The spaces ensure that the pressure is less there, which keeps the metal particles apart and prevents any electrical connection. Otherwise, the contact rows would be short-circuited, which is not the point.

Everything can only work if

1. the pressure is sufficient to make the strips conductive and
2. all contact rows and the pressure plate lie flat on top of each other to avoid contact problems. And that seems to be the weak point.

+++

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

 

[forest bagger]

If these foam strips are stretched or positioned imprecisely, conductive areas will overlap.
If you have conductive silver, Andreas, place the contact foils directly on top of each other and first coat one of the sides that will later touch with conductive silver.

 

[Andreas Thaler]

If these foam strips are stretched or positioned imprecisely, conductive areas will overlap.
If you have conductive silver, Andreas, place the contact foils directly on top of each other and first coat one of the sides that will later touch with conductive silver.

Thanks, Michael!

I'll consider it, but I've already lost too many T90s due to these contact issues.

 

[forest bagger]

Why did you lose them? Did you throw them away?
If not and if there is only this contact problem you may fix it that way.

 

[Andreas Thaler]

Why did you lose them? Did you throw them away?

No, they're used for studies, experiments, and spare parts. Nothing is lost, just the entire camera.

 

[Bushcat]

If these foam strips are stretched or positioned imprecisely, conductive areas will overlap.
If you have conductive silver, Andreas, place the contact foils directly on top of each other and first coat one of the sides that will later touch with conductive silver.

I think I disagree with this: this will bridge the line of contacts, which is not what is needed here.

Andreas: if you ever need to make your own replacement strips, the material is readily available in many forms. A generic strip would have a pitch of around 0.1 mm (conducting slices each separated with an insulating slice). You basically need the pitch of a conducting/non-conducting pair to be at most half the width of a contact, so you can be sure every contact has at least one conducting strip on it, but adjacent contacts are not bridged. In use, you typically need around 25% compression for full conductivity. It's way easier to work with elastomeric carbon/silicone "zebra" strips, but they can't carry much power. I've used both types repairing pocket computers and unnecessarily complex radios. There's also ACF, but you'd probably need a third party to outsource that to: even if you can get the tape in small enough quantities, it's a heat-based curing process.

 

[Andreas Thaler]

I think I disagree with this: this will bridge the line of contacts, which is not what is needed here.

Andreas: if you ever need to make your own replacement strips, the material is readily available in many forms. A generic strip would have a pitch of around 0.1 mm (conducting slices each separated with an insulating slice). You basically need the pitch of a conducting/non-conducting pair to be at most half the width of a contact, so you can be sure every contact has at least one conducting strip on it, but adjacent contacts are not bridged. In use, you typically need around 25% compression for full conductivity. It's way easier to work with elastomeric carbon/silicone "zebra" strips, but they can't carry much power. I've used both types repairing pocket computers and unnecessarily complex radios. There's also ACF, but you'd probably need a third party to outsource that to: even if you can get the tape in small enough quantities, it's a heat-based curing process.

Thank you!

I've only seen these strips with gold-colored particles on the T90. Was this a Canon original solution or standard, also for other devices? Do you have any information about this?

In other SLRs, the contacts are pressed directly together, for example, the Minolta 9000, which also has a large connector and a hard rubber pressure plate. I never had any problems with that.

 

[Bushcat]

Thank you!

I've only seen these strips with gold-colored particles on the T90. Was this a Canon original solution or standard, also for other devices? Do you have any information about this?

In other SLRs, the contacts are pressed directly together, for example, the Minolta 9000, which also has a large connector and a hard rubber pressure plate. I never had any problems with that.

Your T90 connectors may be part # CH2-6005-000, referred to as "gold connectors". I've no idea what that technology is. It's shown in the parts list of "Canon T90 Service Manual Exploded Views and Parts List", part No. C12-1902-000, October 1985. It's typed rather than type-set. It shows all the PCBs and other components laid out. Do you have/want this? I don't know why I have it, I'm not a Canon guy.

The general technologies are ubiquitous. I guess Wikipedia really is easiest source of info on these. Zebra is the grand-daddy of elastomeric connectors:. These use the sliced approach of stacking conductive and insulating layers, hence the name. I've only ever used the carbon ones, but there are also gold and silver variants: https://en.wikipedia.org/wiki/Elastomeric_connector

ACF: typically thin tape on a spool, trim to length. 3M is a major brand but there are many others, including Resonac, the inventor: https://en.wikipedia.org/wiki/Anisotropic_conductive_film

Rather than slices, some types use nickel, gold or palladium-plated plastic spheres embedded in the tape or other carrier medium; they may add a layer of insulation on top of the plating itself, and rely on pressure during assembly to break through the insulation as the spheres rub against each other. I've no idea what type you have in your T90.

 

[Andreas Thaler]

Thanks again, also for the offer document, which I already have.

I know the zebra connectors from the Nikon F3 LCD. Also from the Minolta 7000 main LCD.

The Canon gold version is probably for nerds

 

[Andreas Thaler]

I finished this T90 today and attached the covers and battery compartment. I cleaned some battery electrolyte from the latter and disassembled it for this purpose.

Here are the current results from the camera tester.

The shutter has remained stable compared to the last measurement, and the automatic exposure in Program mode is also satisfactory.

1/4000 s

1/60 s

Brightness distribution of exposure over time.

Exposure error (EE) = -0.4 exposure value (EV).


Everything is very satisfactory.

This T90 can be used to take photos