How to Service Pentax Vertical Run Shutters like the one used in the ‘M’ series of Pentax automatic cameras. The shutter, mirror cage, and body parts are essentially the same for the whole range from the ME, through to the world’s first commercial autofocus SLR, the ME-F.
Here I show one method of checking the shutter speeds and timing, and how to rectify any problems in the shutter itself.
First, let me list the problems normally attributable to the shutter.
- Erratic exposure.
- Uneven exposure top-to-bottom.
- Partially obscured images with flash.
- Dark bands at top or bottom of the image. (Sides for horizontal run shutters.)
- Shutter only operating on ‘X’ or ‘B’ speed settings.
- Unreliable flash triggering.
- Jammed shutter (Rarely).
Note that the shutter itself is only rarely the cause of the shutter jamming, or the camera locking up. (Unjamming ME-Supers is the most common camera repair I am generally asked to do!) This is usually caused by a problem with the mirror mechanism. If there is no obvious damage to the shutter curtains, or a foreign object, or substance (Commonly marmalade!) has got into the camera, then the cause is most likely elsewhere.
If your camera is locked, please look at the ME-Super disassembly page on this site. The problem is common in older examples and usually has a simple cause, and solution.
Also, note that shutter speeds as such are not the responsibility of the shutter itself. Incorrect speeds are due to a problem with the electronics. (Look for cracks and breaks in circuit boards and loose wires.)
Let’s start with checking the shutter operation in situ.
Shutter Checking Equipment.
It is, fortunately, possible to check the shutter and diagnose most shutter problems without and disassembly.
It does, of course, require the use of equipment not normally to be found in the average toolkit, but if you happen to have an oscilloscope or have access to one you can check the shutter operation without much extra equipment.
This method is fairly straightforward and can be applied to any 35mm SLR with a focal plane shutter.
I keep promising myself that someday soon I am going to build myself a compact, self-contained, and inexpensive shutter tester, and when I do I will publish the drawings. Until I do however, this method works well enough for most camera repair situations.
This page does not show you how to calibrate the shutter, you should look at my ME Super calibration page for that. It does show you how to check the shutter and diagnose and fix any problems with the shutter. This is most important, as there is little point in attempting to re-calibrate the shutter if it is not working correctly.
Focal Plane Diode Sensor
Assuming you have access to an oscilloscope, (A dual-trace type, is ideal, but those are quite common these days.) you will need to construct a focal plane sensor.
This can be done with a piece of prototyping board, two photodiodes, and some black velvet or felt. (Or similar black material.)
The one I made looks like this.
From the back
From the front
The back view shows a couple of resistors, of about 100k. These are simply arranged in parallel with the photodiodes and serve to reduce interfering noise. (It also reduces the size of the signal by about 50%, but the ‘hum’ signal is much reduced. Proper screening would probably be a better idea!)
The black ‘felt’ used here is a type sold specifically for use in cameras as an anti-reflection covering. It is essential to use some sort of light absorber in order to get a reliable signal. It is quite surprising how reflective the back of the shutter curtains are. (Especially considering their function!)
In fact, the black felt should cover all of the photodiodes except for a small hole right above the black rectangle of the diode. This helps give sharp edges to the ‘scope trace, making it much easier to read. I haven’t shown it like that here, as you would not be able to see the diodes if I did!
Also, note the orientation of the diodes. The actual active area of the diode is a square. The edges of the square should be parallel to the edges of the board, that helps too. Also, note the spacing of the diodes on your own version. Make sure you measure from the same edges of each diode. That is bottom edge to bottom edge for example. You will need this number when checking the curtain speeds.
I have connected the anodes of the photodiodes together to give a common earth for convenience.
In use this is simply strapped to the film plane, using rubber bands,
and check that the photocells can be seen.
The camera should be fixed in place and illuminated from the front. You could mount it on a tripod, but I have just stuck it to the tabletop with blu-tak. The light source here is a small laser filched from a broken laser pointer with the lens removed, but it works just as well with a halogen desk lamp. (I’m only using the laser because it was just lying about, and it runs for days on a singe battery.)
Connecting the oscilloscope probes now helps find the best position for the light source.
I really should cover the back of the sensor to exclude stray light, but it works OK as it is. Just glue a small plastic box over the back once you know everything is right.
These tests do not calibrate the meter, this has to be done separately, and is dealt with on another page. Once calibrated however, the manual speeds, and calibration can be checked at higher speeds with these tests.
The tests here mainly check that the shutter is properly set-up mechanically. Unless this is done the meter calibration may be of little use!
If you think your meter is in need of calibration check the calibration page first.
The first thing to check is the curtain speeds. With a focal plane shutter, the two curtains are driven by separate springs. The tension of these springs will determine the speed that they travel across the focal plane.
It is important that the curtains maintain a certain minimum speed. If they are too slow, then the curtains may not be open fully when the flashgun is fired. If this is a problem with your camera, then this must be checked first.
It also is important that the two curtains travel at as near exactly the same speed as can be achieved. If they do not, then the exposure will change progressively across the frame. This will be specially marked at higher speeds. If your camera exhibits such symptoms, then the problem lies here.
To check the manual x-sync speed select 125X and set the scope first to trigger from the first photodiode exposed. You should get a trace something like this.
If like me you don’t have a digital storage ‘scope you will probably have to turn the trace brightness all the way up, to get a decent impression of the trace. In this still from a video of the scope, the time base is set at 1ms/division, and we can just about see that the opening curtain is crossing the distance between the two photodiodes in about 4.5ms. The distance between the photodiodes is 18mm, and the manual for the ME-Super states that it should take 6ms or less to cover 21mm, and scaling this up this comes to 5.25ms, so that’s ok.
(If you do not know what the curtain speed should be on another camera, just check what the x-sync speed is. The shutter needs to be fully open when the flash triggers, and so the curtains should cross the whole of the film plane in less than that time. Allow ten to twenty percent less for safety, and use that as the curtain speed. It is important not to set the curtain speed to high, as that risks stressing the springs which could become weakened or even break. With silk shutters, it also controls the shutter tension, too high and the shutter ribbons will tear.)
This should be done several times to ensure that the speed is constant. If it is erratic suspect contamination of the curtain blades with grease or other substance, or that the shutter mechanism needs lubrication. It is also possible that a foreign object has got stuck between the shutter blades. (I have found washers actually inside the shutter curtain guides before!)
With cloth shutters, this would indicate that oil lubricating the rollers need replacing, or that there is some problem with the curtain rollers themselves.
The same operation should then be repeated for the closing curtain. This time the ‘scope’ should be set to trigger on the falling edge, and the trace will appear inverted.
If the camera is calibrated correctly, then the manually selected speeds and the automatic speeds should be within half a stop of each other. The manually selected speeds are only expected to match more closely if the light level is such that the automatic control determines that the precise exposure should be that speed exactly. That is because the automatic speeds change continuously, rather than being selected as the nearest fixed speed.
Check the manual speeds at all settings. If you get speeds greatly in excess of or less than the indicated speed, that there is a problem with the shutter timing circuits. A common cause of this is the crack in the white ceramic PCB, on top of the pentaprism, but you may have to do some deeper investigation to determine the cause.
Shutter speeds greater than a second can be checked using a stop-watch, higher speeds need the oscilloscope.
For this test, you only look at one photodiode at a time. At the higher speeds, you should compare the speeds for each diode.
We start with the ME-Super by checking the x-sync speed, which is a purely mechanical default. I get this trace.
When measuring the times on these traces always measure to the beginning of the change. The time starts when the trace first begins to rise, and ends just when it starts to dip at the other end. This trace is just short of 9ms, say 8.8ms. That corresponds to 1/114s, which is a little slower than the advertised 1/125, which is 8ms exactly. This is just within 10%, and is not a cause for concern given that the curtain speeds are within limits.
The speed should be checked at each speed up to 1/1000s. Check each speed several times to ensure it is consistent. If not, it is likely that the shutter mechanism needs cleaning and lubricating. (Shown later on this page.)
The two blinds should cross the entire focal plane within, (at the worst) 0.1ms of each other. Fine adjustment can be achieved by comparing the shutter speeds at the top and bottom of the run for high shutter speeds.
To do this, select ‘AUTO’ with the batteries removed. This should give you a shutter speed greater than 1/2000s.
At this speed it is important that the two diodes in the sensor board, and the windows letting the light through, are pretty closely matched. The windows less than 0.5mm across, otherwise you won’t get a clear picture of what’s going on.
Remember that the shutter curtains take about 6 milliseconds to cross the focal plane, so that means at 1/1000s the shutters will be separated by 21/6 millimeters apart, or about 2.4mm, at 1/2000 this is 1.2mm. We are trying to get a good look at something of about 1/3000s-1/4000s. If you are having trouble with this, try using the sensor board the other way up, so that the two diodes change places. This will tell you if the differences you are seeing are really due to the shutter or differences between the diodes.
We don’t really care what the separation between the photodiodes is here just that they have a good separation between them.
First, check measure the shutter speed at the top of the shutter travel. This should be fairly regular, it should be pretty much the same for each firing. (Within 10% certainly, but 5% is achievable.) If this is erratic, then the manual shutter speed escapement needs cleaning and lubricating, before you can check the speed matching.
Using an average of several readings of the upper diode shutter speeds as a reference, measure the speed at the bottom of the travel. Again average several readings, and compare the averages.
If the lower diode shows slower speeds, then the closing shutter is running slow. Increase the tension on the upper spring, by one click and re-check the speeds. Stop when the speeds are within 0.03ms of each other. (This may just be a single click on the spring adjuster.)
If the lower diode is faster, then the opening shutter is running slow. Adjust this the same way as the closing shutter.
That is more-or-less all the tests you need to do, the next section shows you the important parts of the shutter, and how to correct any problems that may have been shown up.
Servicing The Shutter
Much of what follows can be performed with the shutter still in the camera, but most of the shutter problems are due to lubrication and dirt, and require the mirror cage to be removed. The ME-Super is not a difficult camera to work on, as it most of the ‘M’ series, and so there is little reason not to disassemble the camera to the point where the shutter can be removed.
The development of shutter faults should be seen as a signal that the camera needs an overhaul in any case, and you should accordingly equip yourself with the required materials and tools to perform a complete service. The page to refer to is ‘Disassembly of the Pentax ME Super Camera’
Overview of the shutter
The shutter is a vertical run metal curtain shutter. It is triggered mechanically, the opening curtain operating completely mechanically. Two mechanical ‘speeds’ are provided 1/125s and ‘bulb’, all other speeds are determined by a retarding solenoid, which holds the closing curtain open for an electronically determined period.
The selection of the operating mode is determined by the position of a lever protruding from the top of the mechanism housing. Viewed from the front of the camera, the leftmost position (Default) selects ‘B’, the rightmost position releases the closing curtain simultaneously with the opening curtain, putting the shutter under solenoid control. An intermediate position provides the x-sync mechanical default speed of approximately 1/125s.
The x-sync contacts are built into the shutter.
Examination and maintenance of the shutter
Starting with the mode selector and default speed rocker.
- Selector’s arm.
- Selector latch. Actually an extension of the selector’s arm.
- Opening curtain latch. This prevents the closing curtain from releasing before the opening curtain is released.
- Mechanical speed rocker. This transfers the motion of the mechanical release cam to the closing curtain release providing the 1/125s mechanical speed.
No oil or grease is used here.
This is the retarding solenoid. When energized it prevents the shutter closing. The solenoid current is switched on just before the mirror reaches its picture taking position by the camera electronics.
This is the solenoid armature. This is pulled into the solenoid when the solenoid is activated.
The mechanical speed cam. Just to the right of the pointer, you can see the cam follower on the end of the mechanical speed rocker. The shutter is in the cocked position ready to fire.
- Is the shutter cocking lever. Moving this in the downward direction in the picture cocks the shutter.
- Is the opening curtain release. This is operated by the mirror actuator as the mirror reaches the taking position. It latches the release cam.
This is the mirror ‘scissor’, not part of the shutter itself, but the spigot just above it engages the cocking lever (Just visible.) during film advance.
The shutter in situ, showing the engagement of the cocking lever with the spigot on the cocking arm.
Indicated is the mirror return striker. It’s coupled to the closing shutter and strikes the mirror return release as the shutter comes to rest.
Circled is the opening curtain latch. The curtain is locked and engaged with the latch. The arrowed part is the opening curtain anti-bounce latch. This prevents the opening curtain bouncing back into the focal plane aperture in conjunction with the opening curtain brake.
In the open position, the opening curtain is latched.
- The old lubricant on this cog and the idler pinions it engages with is the cause of erratic shutter speeds, especially noticeable at high speeds.
- The pin and rubber roller form a brake for the opening curtain which is important in eliminating curtain bounce.
These parts should be thoroughly cleaned with isopropyl alcohol.
It may even be necessary to use a dental probe to gently scrape old dry lubricant from the gears.
After cleaning the gear and associated components should be lubricated with a small quantity of light silicone oil. The type sold in spray cans in motor parts shops can be used. (Spray a quantity into the cap of the can first, and dip probe into the silicone to transfer the lubricant.) Don’t use mineral oil, not even specialist camera or watch oil. This can soak into the nylon parts making them swell, and can make them brittle. Mineral oil will also attack the rubber of the brake.
Note that this gear can be lubricated from the bottom of the camera by just removing the bottom plate.
Also clean and lubricate the cocking arm as indicated below.
Hold the cocking arm in the position shown. Use only the tiniest drop of oil and a light smearing of grease. Failure to lubricate here will cause the cocking lever to stick in this position, jamming the camera.
The screw indicated in the picture below is important and should not be disturbed unless is proves impossible to achieve the maximum shutter speed. It forms part of a contact opened by the opening curtain mechanism and is set to provide a timing reference to the electronics.
The other half of this contact is the gold-plated wire indicated. Dirt or oil on this wire where it makes contact with the set screw could be a cause of erratic speeds or intermittent operation.
The connection to the contact. The brown and the blue wires connect to the solenoid. They can be interchanged, the solenoid does not need to be connected a particular way about.
This ratchet sets the tension of the closing curtain spring. Turn this a few clicks clockwise to increase the speed of the closing curtain. Note that the ratchet only appears to allow increasing the tension. Springs are not likely to become stronger with age.
The lower one adjusts the opening curtain.
This is the x-sync connection. Unsoldering the solenoid and the contactor wire allows the removal of the PCB with the contact intact.
The flash contact here shows some pitting and black deposits caused by arcing. This is due to being used with older flash units which have trigger voltages ranging from fifty to five hundred volts or so. There are a lot of these flash units still in circulation, and this illustrates the sort of long-term damage that can be caused.
In fact, this contact has not seen an awful lot of use and works perfectly well. You can imagine what the result of heavy flashgun use would be like. Devices are available to prevent this from happening but are rather expensive. I have instructions on how to build one for yourself elsewhere on this site and can supply a kit, including a PCB to build it at very much less. (I am looking into the possibility of supplying completed units for those who would prefer it.)
Testing a shutter out of the camera
You may, of course, need to check a shutter without the shutter being in the camera.
You need to make sure that the shutter is held securely in place over the sensor without interfering with the shutter action. Either use the shutter mounting holes or as here only support the shutter at its corners.
The selection arm will have to be held in the ‘manual’ speed positions, central for 1/125s, and to the right for the maximum speed. I usually Blu-tac a matchbox or a rubber eraser to the bench to keep it in place, but usually, I do this with the shutter mounted in the camera body which is easier.
The default operating mode is ‘bulb’, so operating the opening and closing curtains separately is possible. In this case the desk lamp serves as the light source.
That’s all you probably need to know about this shutter. There are plenty of the cameras using it still around, and with a little work, most can be restored to full, faultless operation. Perhaps even better than when they left the factory!