9.5 Miscellany

Home  Contents  9.5  16  Multi-gauge  17.5  28  Pix  Miscellany





Click here for notes an the main pulley for a Pax

A Pathé Cockerel which I have copied and cleaned up 'cos I like the highly stylised look.




An Economy tape splicer for 9.5

Ive been checking and repairing 9.5 films again. Some of you may recall the dicky fit I threw the last time I was doing this; I wrote about it in the Group 9.5 magazine. As before, I am appalled by the careless, even reckless damage I see. Because there will NEVER be any more 9.5 films on anything but the tiniest scale. So if we dont protect what weve got...

This is why I am such a fervent advocate of tape for 9.5 in particular. Cement has its place, but the problem with cement is that you can only join, you cannot repair. Unless of course you use sprocket patches; my advice is:- DON'T! They are the invention of the Devil and it is rare for me to see a sprocket patch that is any good; usually they have involved spreading cement all around the patch, thus ruining the picture instead of saving it. Whats the point? I'd rather cut. Many sprocket patches that I see have dried out anyway (as have many splices) and are ready to fall off. On the other hand, I have experience of tape patches lasting for 40 years without causing a problem. So I use tape extensively, and have found many ways of making runnable a film that would otherwise have to be heavily cut about or even dumped. I don't claim tape will do everything by any means, and it is far from an ideal solution, but it is a valuable tool in the armoury of the 9.5 collector. The problem is, Dr. Leo Catozzo now wants huge sums for his CIR tape splicers, when he can be bothered to make a batch. The last price I paid was £180 and I believe they are much more expensive now. I decided, therefore, to see if I could make an economy tape splicer, but using only readily available materials and tools, so that anyone can do it. You definitely need a file of a decent size, and a set of "rat-tail" files, cheap as chips and readily available on market stalls etc. 

The other really useful thing would be a drill in a stand, but it may be possible to manage even without that (but remember when drilling in metal it must be held tight in a vice for safety and that a spot of oil helps a lot). Other than that, just drills, screwdrivers, nuts and bolts, which the average householder will already have, or stuff that can be gotten from a DIY store.


The big problem with using tape is punching the sprocket holes out afterwards. It is also helpful to have a means of aligning the film and getting the sprocket spacing right when making splices, tho this is less necessary for repair. Anything else, you can wing it. Now, making round holes is easy; making rectangular holes, certainly to the degree of accuracy we require, is much harder. I decided to start with a basic 9.5 splicer of a design which gave me an easy way to get my rectangular hole. I found that the splicer shown in the pictures has locating pins that are held in place by a large blob of solder. It takes time to melt, but eventually you can push the centre pin out, giving you a nice rectangular hole. A bit of cleaning up and judicious enlarging with the tip of a rats tail file and we have a hole of our target size, which is 1mm x 2.4mm. It is important to leave in place the other pins, and the flaps that hold down the film.

The other vital component is a punch, again of about our target size (final size is trial and error by fitting it into the hole). I made this from a strip of steel about 3mm thick and 12mm wide, the only slightly unusual material I used, but which many DIY stores now stock. Here are two views, taken before final filing to finished size. As you can see, I filed from one side only to get to the required thickness, because


9-5miscellany 9-5miscellany 9-5miscellany


it was easier and probably more accurate. Final fitting involved shaving a bit off very carefully and trying it in the hole repeatedly. 

In order to use the splicer effectively, it needs to be fixed down to some sort of board, with provision for holding a roll of tape. Shown above is a tape holder I made earlier, from Meccano. I used nylon rod here, but a bit of broom handle, padded out if necessary with card, will suffice. Incidentally, it is vital to use tape that is completely clear; the CIR splicers used at least to come with a roll of what looked like sellotape and was very cloudy in appearance.

However, fixing the splicer down is tricky as it has no mounting holes. It does, however, have a hole each end. I used a standard DIY angle bracket, cut to size and drilled to take a nut and bolt at the same height as the hole in the end of the splicer. A further nut against the end of the splicer can be tightened to hold it firmly. What you cant see is that I fixed a support under the central hole (with cutaways for the bits of tape punched out to escape) to avoid any risk of bending the bed of the splicer with repeated use.


9-5miscellany 9-5miscellany 9-5miscellany

The rest really is about mounting the punch to an arm. I used a spool arm from a Eumig P8 that had already made the supreme sacrifice by donating its transformer for a 9.5 lighting conversion, plus some lumps of wood and more angle brackets. I found two problems here. First, the arm may be too short or the wrong height or both, and so the punch doesn't enter the hole cleanly, because it is at an angle, travelling along an arc of a circle. The second thing is that it's tricky to align the arm accurately enough and firmly enough. This could no doubt be done, but not without a lot more work. As long as you get a pretty close fit, there is sufficient flex in the arm mounting that you can push and pull to align the punch each time you use it without much problem. The materials used here include a threaded knob I got from Chronos, an unnecessary embellishment probably. I reckon it would be quite satisfactory simply to fit the punch into some sort of holder and just use it in the hand. Why don't you give it a go? I have deliberately kept all this as simple as I can, but there are obviously all sorts of improvements that might be made. It's up to you!


You can see lots of stuff about tape repairs in Care and Repair (if you go to the foot of that page, there is a link to a second page). Below are a few notes that may help if you have no experience of tape splicing. These cover many of the problems with sprocket holes that are encountered on 9.5, particularly where pairs of sprocket holes have one edge weakened, leaving a "chad" or sliver of film that can fall out on projection. Where the chad is still in place, tape reinforcement makes a strong repair and avoids repeated splicing and cutting. This is even more important with sound films, of course.


Using the splicer


1. Cut the film - I use ordinary scissors, leaving more than half of the sprocket hole on each half of the join. This gives a noticeable overlap, which I think is essential. I don't think it is realistically possible to make successful butt joints, especially on shrunken and dried film. The film near the core of a cassette, or even the small cores on some reels, is so twisted it is quite impossible. To quote from the Kodak Book of Film Care "A butt tape splice is considered somewhat superior to an overlap tape joint only [my emphasis] because it is less noticeable (and possibly less noisy) on screen." The Book regards overlap splices as nothing unusual and in my experience they are at least as good as, and no more noticeable than, cement splices.

2. Apply the tape. I mount the tape in front of the splicer and pull tape off its reel using a small-size jeweller's screwdriver under it. This enables me to position the end of the tape at the far edge of the film, leaving no overhang. The screwdriver does not appear to leave a permanent mark (fingers would). The tape is then pressed down and is cut, using a single-sided razor blade or craft knife, along the surface of the splicer, as straight as possible. In time, you will wear a nice groove to follow. I guesstimate where the cut needs to be to give a full wrap-round. (I don't like the way the CIR splicer leaves an edge gap on one side, as tho for a non-existent mag track.)

3. Remove the film from the bed and wrap the tape round, ensuring a close fit to the edge of the film. Trim any excess with a blade or knife; I find it best to keep one especially for this, as cutting on the bed of the splicer does blunt the blade. Increasingly often, I now just use the scissors again with care, the excess can be trimmed without cutting the film.

4. Replace the film in the splicer, close all the film holders and punch. You may have noted that the CIR splicer has a long strip beneath the arm with the cutter and punch; this stops the film lifting up as the punch is raised. In this splicer, the centre film holder performs that task, as the punch passes thru it. I have left the punch completely square, but one could experiment with a slant across the width or even a V-shape of the CIR type.


5. When making joins, any shrinkage of the film is naturally taken up in the join. When making repairs, however, the sprocket holes in shrunken film can be damaged if they have to be forced to fit over both pins on the splicer bed. With my CIR splicer, I have removed two

of the four pins to avoid causing damage in this way. With this economy model, it may sometimes be preferable to use only one of the pins and not close the film holder over the other. Where we are dealing with intact film, there is no alignment problem.

6. I usually apply the tape with one edge following the edge of a sprocket hole (usually one edge is damaged and not the other). The rest of the tape then more than covers the next sprocket hole. In this way, two sprockets can be reinforced at the same time, minimising on-screen visibility as one edge of the tape is masked off in the gate. This can be done working to the left or the right. I try to avoid taping over actual gaps where the sprocket hole has been enlarged, especially if there are several in a row. It looks horrid on the screen and ordinary tape is not designed for this. I sometimes overlap the tape over the offending hole from both sides to give a double layer of tape in an emergency (eg a nearly-vanished notched title). I have also experimented with adding some black felt-tip before wrapping the tape round, in an effort to reduce the on-screen appearance of holes in the film.

7. This technique can be applied to sound films so as to cover only the picture and not the soundtrack, which seems to me sensible. What I do is draw out enough tape to cover the picture and lay it on the film (sound track away from me). I then cut the tape, as in 2 above, but at a different place, in the hope of missing the soundtrack again when the tape is wrapped round. This is OK for repairs but I think would be imprudent for a splice.

8. The other main repair technique I use does not involve the punch. Edge tears can be covered by a short strip of tape wrapped round the edge. The splicer isn't strictly necessary but provides a helpful base to work on.

9. If a cement splice is dried out and coming apart, it can be sufficient to just tape over it as tho making a tape splice. This avoids cutting out any film. Sometimes it is enough to insinuate a bit of extra cement, with the help of capillary action, but if there is damage, eg cracking along a line where the scraper has cut in too deeply, the tape can avoid a cut.




All my headings are more to do with projectors than cameras, so these will have to slot in here. From David Richardson, of course.

9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany

I know you like inside bits, so here is a Baby camera motor, courtesy Tony Reypert.

9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany


A thing that took up a lot of time was the making of a telecine machine.

This involves a major and serious crime that I have constantly inveighed against and for which I can offer only a limited defence. Basically, it involved cutting the side of the lamphouse off a Specto and fitting a camera instead of a lamp. Let me start at the beginning. A year or so ago, Pat Moules and Tony Saffrey saw the machine made by the Welsh National Film Archive to help in the preservation of rare films on 9.5. Instead of the more usual approach, the camera was placed in the lamphouse and a light shone thru the lens barrel. They were fired with a spirit of emulation, as the results were remarkably good. Tony bought a camera and a lens, and muggins here was lumbered with the task of actually making a practical set-up.

9-5miscellanyThe camera had only very limited attachment points - two screws top and 2 more bottom, all at the front end. I therefore made a pair of plates that used these holes to attach them, and left lugs sticking out from the plate at the rear for bolts to hold a rod at each side, which sort of grip the sides of the camera and ensure the plates don't move. You can just see one of the rods lurking at the edge of the black box at the rear, of which more in a bit. (As you can s9-5miscellanyee, there are a series of control switches on one side so this has to be left clear.) To these plates I fixed two angle brackets, to which in turn is fitted a flat plate. This plate slides between guides on the plate in pic 2; the guides are hidden by the brass strips which overlap the plate carrying the camera keep it in situ. The face of the plate is covered in brass sheet, because I found alu on alu did not slide at all well. That's the trouble with what is essentially prototyping, which is basically what I am doing most of the time - you only find 9-5miscellanyproblems as you do things, then it's a choice of start again or find a work-round. The vertical plate is mounted with another angle bracket. You can see at the bottom of the vertical slide that the angle bracket is doubled. This was a late amendment to allow me to use trapped bolts, which you can see sandwiched between the brackets. In the underside view, you can see the big nuts (one smaller cos of lack of space) used for finger tightening, with holes for a tommy bar for full tightening. You can see on the left the extra screw into one of the columns needed to hold the back cover plate on.

9-5miscellany9-5miscellanyHaving gotten to this point, I turned my attention to the projector itself. Having removed the mechanism, and most of the stuff in the base, I fitted the posts you see here. They use the original holes for bolts which held a tranny, except for one I had to relocate as it was partially obscured when the mech was in place. They have a (long) screw thread at the bottom and are secured by nuts from underneath. I9-5miscellany subsequently used the extra length of the threads as the mountings for a plate carrying the 12v power supply. I fitted an extra switch so as to power the camera and the light source separately. These rods support a flat plate on which the camera slide assembly described above sits. Slots in the angle brackets and in this plate allow for to and fro and side to side movement of the camera. Actually this wasn't the first job; that was fitting a plate to cover the shutter to protect both it and fingers.

I next turned to the light source. I Initially made a unit with a 12v miniature bulb I had to hand, but this has now been replaced by a rather more sprauncy LED array, as shown in these pix.

9-5miscellany 9-5miscellany 9-5miscellany 9-5miscellany

The little LED array in the first 2 pix (to give you an idea of scale, they are 5mm LEDs) sits inside a tube (pic 3), which is a nice firm-but-sliding fit in the Specto lens barrel. The other end is covered by a bit of plastic to act as a diffuser. The final pic here shows the underside of the base. The small tranny was originally in one of those bricks you plug straight into a socket, but I cut it open to fit in here. You will also note that I replaced the original Bulgin socket with a standard euro type. Wiring not quite finished yet in this pic and I dunno why that white jump lead is there.

All of this took place as a kind of avoidance activity before I could bring myself actually to bite the bullet and take a saw to the Specto. In the end, I justified it to myself with the thought that I could always make another 9.5 machine by switching parts from a dead 9.5 machine with an 8mm Specto - there seem to be loads of these, in good nick and cheap, and I know it can be done cos I've got one.9-5miscellany

9-5miscellany I finally steeled (aluminiumed?) myself and hacked away the side of the lamphouse so all this stuff I had been making could actually be fitted to the projector. Tony Saffrey brought along a bit of 16mm-reperfed-to-9.5 TV test card and we set it all up. These shots show our very basic lash-up and the amazingly good results we got more or less at once - I spose we didn't ort to have been surprised as the Welsh Archive had done it and all we did was copy. (I hadn't really noticed the shadow under "Pathescope" before).

What we found was that, as advised from Wales, we needed extension tubes between the "C" mount lens (F 1") and the 9-5miscellanycamera. What we also found was that the standard sizes of 5, 10 and 20mm did not quite work - in the nature of things, we needed something between. Fortunately, all the threads on the lens and the tubes have enough length to allow the insertion of "washers" of various thicknesses that I made out of aluminium. We also found that the focussing ring on the lens itself made little difference, tho' it was a bit better at its minimum of about 18". The main impact on focussing was moving the camera assembly itself. However, latest advice from Wales suggests the same can be achieved by an adjuster on the camera which moves the CCD.

This is the thing more or less complete. You can see I made posh knurled knobs and all! At the top you can see a brace I fitted when it became obvious that the angle bracket mounting the vertical slide did not give enough stability. It pivots at the end you can see, and the other end is slotted to cater for movement of the assembly. The slotted end rests on a column, visible past the camera, which took the place of one of the 9-5miscellanyscrews securing the horizontal bed plate. Tony Saffrey expressed concern that the connections at the back of the camera could pose a problem in regular use and be prone to being bashed, so I fitted the black plastic box as a cover. The connections from the camera are led out to sockets mounted on the box, so in general use there is no direct mechanical connection to the camera. If the sockets in the black box cause problems, they can be far more readily replaced than those on the camera itself.

Various bits remain to be done - refitting the arms and the 16mm claw that I removed for some reason yonks ago so it will be a dual gauge machine. I have also fitted a switch on the front below the motor so that the machine can be left plugged in without the tranny being on. I have removed the rubber feet and have metal cross bars to fit in their place so it can all be screwed down to a heavy board for extra stability. (In fact, what I did was to make a sort of plinth, so that the machine can be operated without the front spool having to hang over the edge of a table. It also makes a space where a box for 16mm parts etc can be kept.)

Basically, my part is done and I shall hand it over to Tony and Patrick. I don't actually know how to actually record anything, tho' I did watch an MM M reel all thru. Running a film on a projector and watching it on a TV felt quite odd. Why am I betraying film in this way, you ask? Well, I don't feel that I am. What we are doing is taking advantage of modern technology to extend the use we can make of film. Do you realise that the high-grade frame enlargements in Flickers are copied by digital means? Just another dimension to our endlessly fascinating hobby.

The Pathescope Pax projector (see under Joinville Eqpt) has a belt drive from the motor to a pulley, on which is mounted the shutter. The pulley in UK models is made from Mazac, and has a tendency to distort and ultimately disintegrate, just like the flywheel in a Baby projector. The pulley runs in a narrow slot between lamphouse and projector body, so any distortion causes it to start rubbing9-5miscellany. With the awkward brush arrangement for the speed control contacts, it's all terribly difficult and fiddly to get at. I had to do this once, and get a pulley made (courtesy Buckingham Film Services). I made a drawing for making the new one. It all worked out OK; somebody else might find this of use.

The Mazac problem

Mikael Barnard has found me an article by the late Alan Lott about Mazac. I have no publication reference, but assume it was ACE from the date of Nov 04 on the copy I have. In it, Alan likens our Mazac problems to those of model railway collectors, quoting a piece by Michael Foster in "Hornby Dublo Trains", one of the New Cavendish Hornby Companion series published some years ago. Basically, it seems to be all down to quality control. The basic mix is around 4% Aluminium, 0.04% Magnesium,, 1% Copper and the rest is Zinc. There are two factors which cause the problem: the Zinc must be of high purity, (99.9%) and any contamination of the mix must be avoided. From another source, Alan reported that "modern" standards (basically after WWII ) call for 99.99% pure zinc, with limits of 0.075% on Selenium, 0.0003% on lead, 0.0003% on Cadmium and 0.001% on Tin. These are fantastically tight limits, and it is easy to see howpre-war manufacture of toys and similar products might not have adhered rigorously to anything like such standards. Basically, if the mix was bad or contaminated, sooner or later the item will self-destruct, and nothing can be done to stop it. On the other hand, if all was well, there is no reason to expect any problem. We see this with Pathé Baby shutters - those on the earliest examples are usually fine.

Although Baby shutters are perhaps the best-known example of Mazac failure in our hobby, there are a number of others. The rewind handle for the old KOK seems to have 100% mortality, the Lux suffers in terms of both its gate (early models) and motor end caps. The latter problem also afflicts motors on the Vox/Super Vox and the silent 17.5 Rural. I have recently come across the problem in a Specto casting. New motor end caps can be made, but it's a tricky job unless you're really good at lathe work, especially with the Vox, where gears are mounted integrally with the cap.


A sidelight on splitting of 9.5 films

Leafing idly thru some late 40's ACW's. I came across one comment that made me stop and think. The context was a discussion of the merits of centre versus edge perforations. The comment referred to the way a film gate is "relieved" around the picture area, ie cut back slightly so that, while the edges of the film are gripped to ensure the image is held steadily in the gate, the picture area should not be in contact with any part of the gate except at the aperture, so minimising the risk of scratching. The point was made that, for 9.5 with its central perfs, the film was not directly supported where the claw enters the perf to do its work. In 16 and 8, the sprocket hole is very closely supported by the side runners of the gate.

The suggestion was made that this meant the 9.5 sprocket was more prone to damage and, on reflection, I think there could be something in it. If one imagines gripping a piece of film in a vice, with just an eighth or less of an inch protruding, it will seem very stiff to the touch, and strong enough to cut an incautious finger. A few inches of film would just flop around loose. It would seem logical that, albeit on a very much smaller scale, a similar effect could operate in a projector. There is certainly scope for the film to move in and out fractionally as the claw enters and leaves, which could contribute to splitting. I have often seen 9.5 projectors where wear to the claw has resulted in the underside becoming notched, so that it may actually catch on the bottom of the perf on the out stroke. It would seem possible that the lack of support around the perf could be a contributory factor in any damage. It's interesting as much as anything because I would like to think that there is some reason for the appalling, incessant sprocket damage one sees in so many 9.5 prints other than the sheer incompetence of projectionists.



Converting Cyldon 16mm reels to 9.5

In another typical example of my butterfly mind, I have for no good reason (except that I found it interesting), been looking at co9-5miscellanynverting Cyldon 16mm spools to 9.5. I did two, a 1600 and an 800 and, of course, I found that its harder than I thought. One major tricky bit in reel conversion is ensuring that the sides, and the circular surface on which the film sits, are accurately and concentrically placed in relation to the centre. I have difficulty coming up with ideas to achieve this other than a solid core of the appropriate diameter. Consider the Cyldon reel, which is actually quite clever. There is a central core 2" in diameter. It consists of two discs, pierced for the spool spindle, spaced apart by four pillars, which also pass thru the side cheeks and are then peened over, rather like riveting, to hold everything together. The surface on which the film rests (lets call it the film core) is only attached to the side cheeks, not to the central core. The film core is right next to the central core in the 800 but an inch further out on the 1600. The pix make all this much clearer; this one shows an unaltered 1600'.


The two discs referred to are dished in the centre, and the side cheeks have a large centre hole to fit over this. With this and the separate film core, its not easy to see how to re-use them without a solid central core of the right diameter 2 to 3 inches which would be heavy or wasteful (and expensive) in material by making holes to lighten it. (You should be aware that prices for metal and plastic get very silly when you are trying to buy them in small quantities.) I decided to experiment with tube material to keep the weight (and price) down a bit, and found some in both aluminium and plastic with an inside diameter a bit smaller than 2" that I could bore out for a tight fit round the central core, which I planned to re-use to circumvent the sort of problems outlined above. I drilled out the "rivets" on each side of the core with a drill about 4.8mm. This size is about right to separate off the part of the riveting above the surface; dont drill too deep because we want to re-use the pillars. Once the pillars are released in this way, you have to prise off the film core; I usually try to get a screwdriver blade between film core and spool cheek and twist to force them apart. Damaging the film core is OK, but not the spool cheeks. If you can't get started, I have found that if you stand the reel vertically, you can hit the film core very hard without damaging or distorting the sides. I use a cold chisel, but a heavy screwdriver would probably do. Once started, its easy to follow the gap around to prise the rest off.


In re-assembling, I gave very careful thought to appearance, mindful of those otherwise useful giant 8mm spools that were marred by huge nuts and bolts. I decided to go for 3mm screws, my default size for a great many things. The heads are small enough to be unobtrusive, particularly with the flat, round-headed type. I originally wanted to use countersunk heads, which could fit pretty much flush with the surface of the spool cheeks, but this didn't work. The reason was that, in order to maintain accurate location, the adapted pillars had to extend thru both the discs of the central core and the side cheeks, as a sort of registration pin. In the original, the rivets themselves were the right size (about 5mm), but 4 x 3mm screws thru holes in both discs and cheeks would leave a lot of room for slop and could easily leave the reel unbalanced. Trouble is, the side cheeks are only a bit over 1mm thick and there is no room for both pillar and the taper of a countersunk head. I found the screws were not able to grip the sides firmly enough to the pillars because they tightened against the pillars first. 


The9-5miscellany new pillars are made from the old. They need to be trimmed down to a central length of 8.5mm approx, with 2mm extensions at each end, 9-5miscellanytrimmed down to about 4.7mm diameter. A hole is bored thru the centre and tapped with a 3mm metric thread. The picture left shows the component parts of the 800' conversion, first assembled then apart.. For the spindle core I followed exactly the pattern of those used on the 9.5mm 900 fibre reels. This fits neatly into the 16mm holes in the central core. Note only 3 of the 4 pillars are visible - one is still in one of the discs. 

For the 1600 spool, I used black acetal (a kind of nylon, I think), in the form of a tube of about 3" external diameter and 1.75 internal. I bored this out to make a tight fit round the central core I didnt really want to add more screws, tho I suspect I may have to. It still falls well inside where the original was, so I made sure to smooth off the holes where the film core had been fixed, to avoid snagging the film. I normally attach film to spools by friction alone, especially those spools with larger cores. However, I did make a couple of token saw cuts at an angle into the plastic that could take an end of film.

For the 800, I used aluminium, with a much thinner wall. The 800 original has cut-outs in the spool cheeks to enable film to be threaded into a slit in the film core. I decided to provide such a slit but, as I was not planning to secure the film core other than by friction, I thought it best to put a strut across a chord of the film core before cutting the slit (seen above left). In retrospect, I think it might be as well to fix the film core with screws these could go thru the holes left by the original, tho they would have to be small to fit into the thin metal of the film core. The actual width of the film core is about 11.5. This seems to me a bit wide but it matches the old Pathescope ones and, with a straight-sided spool, it may be best to have a bit of leeway. The only remaining task is to drill holes in the right place to take the drive pin on the projector spindle.

Finally, here are the completed articles.


9-5miscellany 9-5miscellany




Home  Contents  9.5  Multi-gauge 17.5   28  Pix  Miscellany