8mm DIY Telecine

Film transport

Below is a picture of the modified projector internals.

An aim here was not to modify the projector more than was necessary, and to do nothing if possible that would stop it being used as a projector at some point in the future. In fact the only physical modification is two 3mm holes for screws which can be seen at the top of the case to hold opto sensors, two small cutaways in the plasic cover to allow the sensor and lamp cables through (on left) and two holes in the front to mount the film cleaning mechanism. The rest is done by removing and replacing parts.

The fan shroud has gone (very little heat generated without the tungsten lamp - see later), although the fan is left in place. That is simply because the air viscosity provides some load on the motor that widens the speed control range. The speed control potentiometer can be seen bottom left - warning: that has mains AC voltage on it.

I haven't looked into the motor wiring, but assume the transformer (a) acts as an auto-transformer to provide the different mains voltage selections (I noted the motor slowed up when the lamp was switched on) and (b) provides the low voltage for the 50W tungsten lamp. The motor can be seen bottom right with the drive belt to the right of the fan. The shaft above it drives everything else.

The film transport (not shown - on the other side) consists of a small pulley (toothless) that pulls the film through, and a pulldown claw that advances it a frame at a time. This in turn is driven from a cam, so that during one rotation of the drive shaft (geared down from the motor by 3:1), the film is only being advanced for about 10% of the time, and for the rest it sits still in the gate. This cam and claw driving mechanism can be seen to the right of the worm gear on the left. Because the pulley is continuous drive, and the claw performs a sudden pulldown, the film has a spring tensioned loop between the two.

There is a three blade shutter on the drive shaft which uncovers the (stationary) film three times per frame movement.

That was the first thing to go. As was seen, the decision had been made to use a CMOS rolling shutter camera, and that means that the image is being captured for most of the film's 'stationary' period. If the shutter was left in place, black bars would appear across each frame.

It was replaced by a similar disk, made of half hard alloy, that still had three slots but for which the outside diameter was sufficiently small that the film is permanently uncovered. The three slots are used to trigger a reflective opto-sensor, which feeds the electronics.

One of the slots has a further inner slot that triggers a second opto-sensor to give overall positioning information (ie which slot is which).

The sensors (OPB608B) can be seen as the black objects mounted on the left side of the paxolin stripboard opposite the slots.

The outer sensor triggered by the three slots provides an interrupt to the microprocessor (a PIC) controlling everything, and the inner then gives it a reference as to which is 'slot 0'.

The disk is positioned so the triggering edge of the 'outer slot 0' occurs about 45 degrees after the film has stopped moving following its pulldown, although the actual position is not crucial as adjustment is carried out by software.

The time between outer slots gives a measurement of frames per second and a second timer triggered automatically from one of the slots causes the camera to be triggered after a time delay (dependant on the measured speed) that brings it to just after the film movement phase, and thus allows maximum time for exposure and data transport before the film starts moving again.

The microprocessor runs a simple user interface with some switches, buttons and a 16x2 character display, and one thing it allows is moving that trigger point around to get it in the right place.

Film cleaning and lubrication

This was a fairly late addition.

Some of the films were in a sorry state (one had mud on it). Even with the cleaner ones there was always the problem of small hairs etc sticking to the gate, and some did not want to feed properly at all. I also noticed a lot of dirt building up in the pull down mechanism pressure channel. I concluded that some form of film cleaning would be needed.

Some internet research indicated that film lubrication might also be needed, and there are a lot of articles out there with different opinions on the best way to do it, and the chemicals to use - Filmrenew and Filmguard coming up a lot (with both positive and negative comments), though they seem to be hard to come by in the UK.

In the end, rightly or wrongly, I took a simpler approach based on Gerald Santana's suggestions on the 8mm forum. I mixed some cleaning fluid from a mixture of Isopropyl Alcohol (as pure as I could get it to reduce water content) and mineral (or paraffin) oil, in a ratio 1000:1. Rather to my surprise the oil does appear to dissolve in the alcohol. Is this a good idea? Don't know. If I take out a film one day and all the emulsion has disappeared, I'll know it wasn't, but so far it seems to be doing its job.

For particularly suspect looking films I first run them between hand winders (actually a dual 8 Goko film editor) with the film running between pads (pec pads) which have been wetted with the fluid. Then for all films I have added some pads to the incoming film path on the projector, again wetted with the fluid (on fresh pad surface) before each run.

This helped in a number of ways: there are less dust spots on the images, there is less likelyhood of hairs in the gate and the film feed problems appear much reduced. It also seems to have helped with the wave problem mentioned later.

There is now an 8mm version of the 35mm automatic film cleaner commonly used professionally created by Roy Neil (see 'Film O-Clean'), where the film runs over rollers, which then drive feed spools of cleaning material (well geared down) and so continually moves fresh cleaning material into the film path. This appears to be a most elegant solution, but at the moment a little too expensive for me.