WOT: Keystone A-82 16mm Projector

johnmeyer wrote on 6/2/2011, 3:45 PM

OK, found what he said but the two frame rates he suggested are for 8mm and I'm assuming by the title of this thread that you're working with 16mm which is nominally 24 frames per second. I'd suggest starting with 24 rather than 16 or 18.Well, I never thought I'd disagree with anything Bob said about film because he spent time in that industry and I didn't. However, if your 16mm film is silent, then it will almost certainly have been shot at something less than 24 fps. Most amateur 16mm film was not shot at 24 fps. In particular, if your film was shot during the same era as your Keystone projector was manufactured, then I can guarantee that it was shot at a much slower rate.

Almost all of my transfers from the 1930s seem to be at about 16 fps. I have set the playback for some as low as 12 fps.

I have now set up a "playlist" on YouTube that has all of the film transfers I've posted, and they are in chronological order. The 16mm transfers from 1928 are timed to playback at 12 fps.

John Meyer's Film Transfers

As for answering the question about framerates, let me reiterate and expand on what I posted above.

The playback speed of the projector, as I already stated, doesn't matter IF you eventually -- through software or other means -- end up with a "frame accurate" transfer, meaning that you have exactly one frame of film on each frame of video. Once you have a frame-accurate transfer, you then have to set the playback rate. If you are simply going to play the video file on your computer, or upload it to YouTube (or some other web site), then all you have to do is set the playback flag in the AVI file to whatever frame rate you want. I use a simple hack utility called "AVIFrate" that will set the fps header on any AVI file. It takes a fraction of a second to run, because it simply changes the fps number in the header.

However, if you want to put your film transfer on a DVD, it is slightly more complicated. To create video for rendering to DVD, you put the results of your frame-accurate transfer back on the Vegas timeline, disable resample for this media, and also set it to progressive. You then set the Vegas project properties to progressive, and also change the project properties frame rate to match the frame rate you set (using AVIFrate), e.g., 16 fps. When you then render to 25 fps (PAL) or 29.97 fps (NTSC), Vegas will create a more-or-less standard pulldown, without blending frames (because you disabled resample) and you should get a very nice, crisp playback, at the correct speed.

As for what frame rate to use (12, 15, 16, 18, or 24), you simply have to watch the video for awhile, after setting the playback flag, and decide if it looks too fast or too slow. Sometimes you can use a trick. For instance, if someone drops something from a height that you can estimate, you can use that as a way to determine the frame rate using basic physics: t = sqrt (2d/g). If you are lucky enough to have a clock with a sweep second hand in the frame, you can get very accurate with your estimates.

Of course, once you figure it out, you'll find that it may be different for the next reel. Many amateurs changed cameras, and all cameras were different. Some cameras used wind-up springs, and the governors on these would sometimes speed up or slow down as the spring wound down. And, of course, the really early cameras were hand-cranked, although not many of us get a chance to transfer footage that old.

amendegw wrote on 6/2/2011, 4:22 PM

I love this stuff!!!

I wish I still had my Dad's camera. Here's what I remember for sure: it was 16mm, spring wound, took Kodak 16mm cassettes and only had a single lens. It was purchased in 1948.

I seem to recall it was a Bell & Howell - but can't be sure. I Googled this and didn't see a camera that fit my memory.

btw, I got the notice that my new projector bulb was shipped today. Maybe more testing in a week or two.

...Jerry

Edit: johnmeyer, it's a real treasure to have your Dad narrate the old film clips. Hopefully your children & grandchildren will value this!

johnmeyer wrote on 6/2/2011, 8:27 PM

it's a real treasure to have your Dad narrate the old film clips. Hopefully your children & grandchildren will value this! Just yesterday I briefly helped someone from Paris, over at the doom9.org forum who is transferring old film for his dad. I recommended he get his dad to talk while he watched the rough cut, and then add that audio. He didn't quite "get" the idea until he heard those transfers where I included dad's narration, but when he heard it, he got quite excited and I think is going to do it.

By contrast, after I transferred about forty 400' reels for my brother's father-in-law, I gave the same advice to my brother, but for eight years he kept never quite getting around to it, and the old guy finally died this winter at 92. He had all his marbles up until the end but, at least as it relates to the three miles of film, he took his marbles and went home without letting us in on his game.

So yes, everyone, including the grandkids, will appreciate it, and I sure get a kick every time I listen to any of it.

Serena wrote on 6/2/2011, 10:21 PM

If it took Kodak film magazines I suspect it was a Kodak camera, although not necessarily: http://www.agassiztrading.com/cameras-film/cameras/cameras-16mm/16mm-cine-kodak-magazine16.htmKodak magazine loading 16mm[/link]

farss wrote on 6/3/2011, 1:56 AM

" Most amateur 16mm film was not shot at 24 fps. In particular, if your film was shot during the same era as your Keystone projector was manufactured, then I can guarantee that it was shot at a much slower rate. "

It seems it was shot in 1949 according to Jerry. It's in colour as well so it'd have to be quite some time after the 1930s.

I did say "nominal", as I've transferred regular 8 that was shot at 24fps from around the same era, admittedly by a quite serious amateur doco maker. On the other hand I never came across 16mm from before the 1950s so I'd say you're right, just about any frame rate would have been the go back then.

Bob.

amendegw wrote on 6/3/2011, 2:37 AM

Serena,

Thanks so much for that link. Browsing thru those pictures, now I'm almost sure the camera was a Revere Magazine 16, not a Bell & Howell. I'm not sure this is the exact model, but it's very close. Does that give us any clue regarding the frame rate?

Wouldn't all these "magazine" cameras have the same framerate? Or is that a bad assumption?

...Jerry

farss wrote on 6/3/2011, 6:02 AM

"Wouldn't all these "magazine" cameras have the same framerate? Or is that a bad assumption?"

I wouldn't lose too much sleep over that. As John very rightly said the most important thing is to get a frame to frame transfer. Once you have that you can then change the frame rate in the digital realm.
I suspect you're right, those "magazine" cameras probably would have had a fixed frame rate. Certainly the one you posted a photo of seems to have no way to change frame rate. Keep in mind they were clockwork so the frame rate wasn't terribly precise but it doesn't matter. I doubt many can tell a difference of even 10% in frame rate, come to think of it +/- 20% would probably not be that noticeable. It was only the "talkies" that made fps important, a 10% change in pitch IS very noticeable.

BTW, you've now made me dig out a Eumig C3R double 8 camera someone gave me years ago. That over 50 year old thing still works, even the exposure meter. The lenses need a good clean. That camera will run from 0 fps to 32 fps but it does have "16" engraved in red.
A Google found me a supplier of 100' reels of 100 ASA Ektachrome for the camera. Very tempted...

Bob.

Serena wrote on 6/3/2011, 6:13 AM

We're talking about mechanical cameras and their frame rate was determined by a mechanical governor. If it was single speed (which would be unusual) it would have been set for a nominal 16fps. Otherwise the frame rate was set by a dial marked in frames/sec, but this wasn't click indexed so could easily be off by a frame or two per second. Also, as John mentioned, the torque of the clockwork motor varied as it ran down and governors varied in effectiveness. The silent speed of 18fps came about because that was the minimum speed at which B&H Filmosound projectors could raise their safety shutter; it was lifted by fan air pressure. These projectors were very common, so amateurs shot at 18fps. In fact very few saw any point in increasing film costs by 50% to shoot at sound speed (unless there was some hope of commercial release); 18fps gave quite satisfactory results.
Recently I was reminded of those times when somebody was critcizing a new video camera as being no good for run and gun because something wasn't automatic. People have got very soft! WWII was run n gun shooting with wind up cameras -- often B&H Filmos. Manual focus (no reflex viewfinder), manual iris (exposure meter -- had to be better than 1/2 stop if shooting reversal), wind up the camera after each shot, perhaps only 15 seconds on a wind, 3.5 minutes for 100ft of film, manual reload. Magazines were a boon for fast reloading.

Serena wrote on 6/3/2011, 6:20 AM

>>>>>100' reels of 100 ASA Ektachrome for the camera. Very tempted...<<<<<

Bob, you don't won't 100ASA unless that is all that's available -- you want fine grain and Kodachrome II preferably (but no longer processed by anyone anywhere). I think 40ASA Ektachrome might still be available. And that C3R takes 25ft spools of double 8.

EDIT: oh yes, you notice 10% difference between camera and projector.

johnmeyer wrote on 6/3/2011, 10:16 AM

Someone posted a link (over at doom9.org) to a very good overview of how to transfer film to video:

Restoring 8mm Home Movies

amendegw wrote on 6/26/2011, 8:10 AM

Okay, I've been working on this project on-and-off for the past month and frankly I'm getting discouraged. I thought I'd make one last post here to see if anyone had any ideas (or maybe convince me to abandon the effort an go to a commercial service). Here's some random observations:

1) The use of the original 750w bulb was very difficult because it was just too hot & bright. And... the transport mechanism on the projector is old-and-clunky and I'm having difficulty operating it while keeping the bulb cool.
2) The Eiko 100W bulb recommended by johnmeyer was (almost) perfect. Not too hot & bright. The image on my capture device (see below) was sufficiently bright. I could freeze frames to focus & align without fear of burning the film. However, the first bulb blew out within the first 5 minutes of use. Figuring it was a bad bulb, I RMAed it. Next bulb came in last week and it, too, blew within 5 minutes (must be my projector & not be the bulb).
3) I tried jury-rigging up a small 40w chandelier bulb that would fit in the bulb enclosure, but discovered that the filament geometry created hot-spots in the image and the 40w was too dim to get an adequate capture.
4) I can try to rig up an alternate light source, but would like some ideas. Ideally, it would be a 20mm diameter, evenly distributed, cool, and about the same brightness as the 100W tungsten. Any ideas?

5) One more thing, I've found that it is much easier to use the following contraption, rather than the direct into-the-projector-lens method as I was having too many problems with alignment and focus.

TIA,
...Jerry

Edit & PS: I took the above stills with the still capture feature of my Panasonic TM700 and was pleasantly surprised with the quality of the stills.

johnmeyer wrote on 6/26/2011, 9:00 AM

Wow, burning out a bulb that is 1/7 the power of the original sure isn't something I would have expected. I can think of three things that might cause that:

1. Excessive vibration. If there was something worn, I suppose the bulb could be exposed to unusual amounts of vibration. Even without a defect, perhaps the projector bulbs have a rugged filament that this bulb doesn't have.

2. No cooling. There should be a fan somewhere, pushing air past the bulb. If that isn't working, any bulb is going to overheat.

3. Voltage. Most bulbs are 120V, and I think I linked to a 120V bulb. However, if you used a 12V bulb or 48V bulb, or anything less than 120V, it would fail quickly.

amendegw wrote on 6/26/2011, 11:14 AM

Not to belabor this issue because the chances of two consecutively blown bulbs is pretty conclusive evidence that something is wrong with my projector or its pairing with this bulb.

That said, I got out my 35 year old Micronta analog mutlitester (one of the handiest things I have in my toolkit - even though I only use it once every 6 months or so) and the lamp socket tests dead-on at 120V. Not surprising because I didn't see any transformers on the projector.

The specs of the bulb are, indeed, 120V.

I can't run the projector without the fan coming on (and it seems to move a lot of air).

Vibration is a possibility, the A-82 does not run particularly smooth, but it also doesn't seem to shake enough to blow the bulb within 5 min. And the original 750w bulb doesn't blow (but the filament is much more substantial).

Who knows?

...Jerry

PS: Who's a EE? I seem to remember from a EE course I had in college that circuit with a resistor & capacitor (maybe a transducer had to be in there, too - my course was 45 years ago) will cause the amps in the circuit increase slowly when switched on. Could it be that the lower resistance of this new bulb is causing the power to spike?

JJKizak wrote on 6/26/2011, 12:15 PM

Having shown hundreds of 16mm films bulb blowout is very common. There are times when a bulb will blow out only after a few seconds of operation. Make sure you do not leave fingerprints or sweat on the bulb, and keep plenty of spares handy. Some of the bulbs will last forever, well it seems like forever. Make sure the cover is on so that the cooling air is directed properly.
JJK

johnmeyer wrote on 6/26/2011, 2:46 PM

Who's a EE? I seem to remember from a EE course I had in college that circuit with a resistor & capacitor (maybe a transducer had to be in there, too - my course was 45 years ago) will cause the amps in the circuit increase slowly when switched on. Could it be that the lower resistance of this new bulb is causing the power to spike? I'm one of several EEs that post here. There is definitely an in-rush current into any incandescent bulb. However, most bulbs are just connected to the main power line, without anything in between, and this would especially be true with an old projector like this. It should be able to withstand this, and if there were any issue like this, you'd see it immediately at turn-on time (it would pop when you turned it on).

Having said that, it is possible, I suppose, that the bulb might be wired in series with the motor. That could cause some pretty strange things to happen. That seems unlikely, however.

Another possibility is that it is indeed wired directly to the main circuit, but that the old motor is causing some sort of voltage spikes as it spins. Since this is a lower-wattage bulb, it has a higher resistance and would be more susceptible to these spikes than would the normal high wattage bulb. My reasoning is that the high wattage bulb is closer to being a short circuit, and this would dampen the spikes.

I don't have a lot of conviction behind this analysis, but if it is correct, the solution would be to put a large capacitor directly across the leads of the lamp. You need a capacitor rated at 250V and which is bi-directional (i.e., NOT a polarized electrolytic). These are commonly found at the input stage of most switching power supplies. They are usually relatively small (compared to electrolyctics), typically about 0.1 uF.

The better alternative is to go the route that you already started, namely to figure out how to fit some other el-cheapo bulb into the same place. Roger Evans did this in his original Workprinter modifed projector models. He used a small-base round frosted bulb, like this one:

40 Watt, 120 Volt G16 White Long Life Globe Bulb

The large size of the bulb negated the hot spot problem.

In my projector modification, I couldn't do that (it wouldn't fit), so I used a lower-wattage bulb, but then mounted a piece of frosted glass between the bulb and the condenser system. This completely eliminated the hot spot. You have to make sure to use a fairly small-wattage bulb if you do this, or you may shatter the glass (unless you are able to get heat-resistant frosted glass, which is a pain to obtain).

amendegw wrote on 6/26/2011, 3:28 PM

"In my projector modification, I couldn't do that (it wouldn't fit), so I used a lower-wattage bulb, but then mounted a piece of frosted glass between the bulb and the condenser system. This completely eliminated the hot spot"First, if I ever forget to thank you for going "above and beyond" in explaining and researching a project that may never pan out, please let me say "Thank you!!" now.

Next, I mentioned my "jury rigged" system that I came up. Here's a picture of what I came up with:

I didn't seem to have overheating problems as the wattage was low (40w) and the fan (designed for 750w) worked pretty well on this little guy. As you may deduce from the photo, the electrical source is completely separate from the projector. The problems where... the 40w was not quite bright enough and the filament caused hot spots.

I think I'll look around and see if I can't find a >40w frosted bulb with a standard candelabra base.

...Jerry

Edit: Maybe this: Satco S4485 -- 60 Watt - Xenon/Krypton
or maybe this: Feit Electric Mini Candelabra

farss wrote on 6/26/2011, 3:33 PM

The resistance of a tungsten lamp cold is lower when it's hot. Also the filament is a coil and the inrush current can cause the filament to shake and fail. The lower the wattage of the lamp the more fragile it is. Higher resistance dictates thinner wire.

One simple explaination for why your lamps are failing is they are very old stock. A lamp will die very quickly if there's any oxygen in the lamp. Glass generally is slightly porous so over the years some air leaks in. Projector lamps I think were made with positive pressure from an inert gas, the opposite to conventional lamps.
You *might* get around the problem by running the lamp from a variac, do not use a conventional dimmer, no sir, they will make matters worse.
Run the lamp for a few hours so it is just glowing and slowly increase the voltage to spec or just under.

Bob.

amendegw wrote on 6/26/2011, 3:44 PM

Bob,

I'm sure you speak wisely here, however, I've already gone thru 2 bulbs at less than 5 minutes each. I'm not sure I want to buy a stock of $30 bulbs (incl shipping) and then a variac - to experiment with the best way to "prep" them.

I'm not abandoning this project yet, but I'm moving closer to using a commercial service.

...Jerry

farss wrote on 6/26/2011, 4:12 PM

Indeed you're probably on a road to nowehere buying more of the same lamps. You might do better with a 50W 12V halogen dichroic lamp. You can buy them, a socket and the transformer very cheaply at any lighting retailer. Because they produce a narrow focuseed beam they may work quite well.
A bit of simple metalwork is going to be required though to hold the lamp in the right place. I have a couple of old "retort stands" from my high school days, great for holding all manner of things in place while I see what's going to work before I start cutting aluminium.

Bob.

amendegw wrote on 6/26/2011, 4:46 PM

Bob,

I think we have been posting concurrently. Go four posts up and look and my jury rigged contraption. Unless there is a better idea out there, I think my next step is to get a high wattage halogen or Krypton/Xenon frosted candelabra base bulb.

...Jerry

Edit: I just ordered a ""Satco Light Bulbs 100Q/F/JD 100watts E11/T4 Mini Can Halogen Bulb, Frosted"" from amazon.com for $7.79 w/ free shipping. Worth a shot.

farss wrote on 6/27/2011, 12:29 AM

Jerry,
sorry mate, I missed that one.
Let us know how that lamp works out. I'd be just a tiny amount concerned about running the lamp base up, just watch out for things getting too hot although I think you said there's a fair amount of airflow so you should be OK.

Bob.