Film-Tech Forum ARCHIVE: Cement Splices

This is topic Cement Splices in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=000714

Posted by Joshua Lott (Member # 240) on 03-04-2000, 12:36 AM:

I was wondering what the deal is with cement splices? You can get a print that doesn't have a single one, and others that have 1 or 2 in every reel(Next Best Thing). I have also heard these refered to as 'factory splices'. What is the correct term?

Posted by Brad Miller (Member # 2) on 03-04-2000, 02:21 AM:

Since all release prints today are on polyester stock, "technically" those are called ultrasonic splices. I've always called them lab splices. They are where the labs splice the end of rolls together to save on print stock. I'm sure John Pytlak could explain this quite well next time he's perusing the forum.

Posted by John Pytlak (Member # 331) on 03-04-2000, 03:21 PM:

Unfortunately, "lab splices" are an economic necessity in release prints. Film manufacturers usually supply rawstock print film to the labs in 4000-foot and 6000-foot lengths. Since release print reels vary in length, the labs would have to discard "short ends" if they had to produce splice-free prints. (e.g., if a reel is 1800 feet long, labs would have 600 feet of "short end" waste if they printed three splice-free reels on a 6000 foot roll, and discarded the rest).

With polyester print film, labs use ultrasonic splices. The big Hollywood labs often use high speed continuous feed-on printers, that use an "elevator" to allow time to make a splice without stopping the printer. Other labs usually use slower bi-directional contact printers, and make the splices off-line. Since the splice is made in the dark on the film before it goes through the printer, the splice may not always fall on the frameline. Also, the undeveloped emulsion trapped in the ultrasonic weld, and the heating of the film to make the splice, causes some discoloration at the splice.

Lab splices that are not on the frameline, or that are very visible, should be properly remade by the projectionist during print inspection and make-up. Generally, ultrasonic splices are very reliable, but they sometimes aren't very pretty on the screen. So remake them if needed.

------------------
John P. Pytlak, Senior Technical Specialist
Worldwide Technical Services, Professional Motion Imaging
Eastman Kodak Company
Research Labs, Building 69, Room 7419
Rochester, New York, 14650-1922 USA
Tel: 716-477-5325 Fax: 716-722-7243
E-Mail: john.pytlak@kodak.com

Posted by Brad Miller (Member # 2) on 03-04-2000, 03:28 PM:

John, how about making little holes every 4 perfs on the picture edge of the film between the sprockets similar to 70mm? Think there's any chance the labs would actually use those as a reference to ensure all lab splices were on the frameline?

Posted by John Pytlak (Member # 331) on 03-04-2000, 03:40 PM:

Many times, even 70mm was not printed in reference to the round holes every fifth perf, so I don't know why 35mm would fare any better. Remember, those lab splices are made in the dark, and often have to be made within about 20 seconds.

Anyway, Kodak eliminated the reference holes on 70mm print film some time ago. They were a potential source of dirt, since it was difficult to maintain the "sharpness" of the perforator punch and die for such a small hole.

Posted by Scott Norwood (Member # 30) on 03-04-2000, 05:00 PM:

Question for Mr. Pytlak - if lab stock is distributed on 4000' or 6000' lengths (which is my understanding and which is what you state above), then what is the explanation for the occasional reel that has more than one lab splice in it? It would seem that it would be impossible to end up with more than one lab splice per 2000' reel, but I've seen it happen. (And, yes, this was with a lab-new print...it wasn't a "composite" print made by salvaging the good footage from several old prints.)

Also, it seems that lab splices are more common now than in the past. With the exception of "old" IB Tech prints (which seem to always have a lab splice in each reel), none of the "old" (pre-1990 or so) feature prints in my (admittedly small) 35mm collection have lab splices. Same for the older prints that I've run theatrically (though some of them had lots of tape splices, some of which might have been made to replace lab splices). Maybe I'm just lucky. Who knows?

Posted by John Pytlak (Member # 331) on 03-04-2000, 06:23 PM:

The ACVL (Association of Cinema and Video Laboratories) Handbook notes: "In release printing, it is recommended that there be a maximum of two splices per 400 feet of 16mm or 1000 feet of 35mm. These two splices may be insert splices, or one of the two splices can be a raw stock splice. There should never be two raw stock splices in the same reel. It is further recommended that spliced raw stock not be used in trial prints, dailies, TV spots or recording prints. It is customary practice to charge a premium for release prints that have no raw stock splices."

Posted by John Pytlak (Member # 331) on 03-04-2000, 06:29 PM:

Scott:

To further answer your question, if you find more than one raw stock splice in a single 2000-foot reel, the lab may have spliced several short ends together off line. As you can see, this is not in accord with the published ACVL practice. You are correct that the labs that use continuous feed-on printers would normally only have one lab splice every third reel, assuming nominal 6000-foot raw stock rolls and 2000-foot reels.

Posted by Christopher Seo (Member # 7) on 03-04-2000, 07:19 PM:

I thought that the ultrasonic splicers that have sometimes been mentioned in this forum are supposed to produce very-low-overlap, neat splices whereas the lab splices I've seen are just the opposite. Do lab splicers simply not have enough time to do the best job?

Posted by Joshua Lott (Member # 240) on 03-04-2000, 08:20 PM:

I always replace factory (sonic) splices because even if they are on a frame line the film still tends to jump a bit. I also make it a habit to replace old splices on a used print when we get them in because it seems like 9 times out of 10 I get stuck with the yellow zebra tape that is applied wrong. Hence flashes of yellow at reel changes.

Posted by Robb Johnston (Member # 239) on 03-05-2000, 02:50 AM:

I splice out as many of the sonic splices as I can as I had one come apart 6 weeks or so into the run of a film about some big ship and an Iceburg. Anyway, I am more annoyed at the fact that more often than not I am finding these splices within 5 feet of the end of the reel, and this last week even in the leader which seemed ever so unnecessary. The all time classic had to be the one version of A Bug's Life Trailer which routinely had a factory splice in it.

Posted by Randy Stankey (Member # 64) on 03-06-2000, 01:38 PM:

I saw a trailer that had TWO lab splices in it! (The ultra-long version of Titanic) I just tossed it out!

I have a personal rule that trailers and policy strips, etc. are allowed to have ONE splice in them. (Assuming the rest of the trailer is pretty much perfect.) Any more than that and they become Christmas tree decorations!

Posted by Masao Garcia (Member # 184) on 03-06-2000, 01:46 PM:

I just wanna add that I don't take out lab splices unless they are somewhere on the visible frame, because tape splices tend to attract dirt over time and the image quality tends to degrade after a while, where in the lab splice case, there is no added attraction of dirt. This pertains only to those of us not fortunate enough to be allowed to use FG. But I can see everyone's points.

Posted by Joe Redifer (Member # 3) on 03-06-2000, 03:19 PM:

I usually don't check for lab splices in trailers. But once when I built up Star Wars Episode 1 the DTS trailer had a lab splice in it! I never removed it. Eventually the trailer was tossed exactly 12 weeks later when we finaly lost the print.

What's next? Lab splices in leaders?

Posted by Scott Norwood (Member # 30) on 03-06-2000, 05:48 PM:

As for lab splices in leaders-I have seen them. I have here a "new" IB Tech reel of GWTW with a lab splice in the countdown. This is an interesting situation, since removing the splice would require replacing the lost footage with a "slug" of black leader to maintain the proper footage count. Personally, I wouldn't bother to remove a lab splice in a leader unless it fell apart on its own accord (in which case I'd just replace the whole leader).

Posted by Tim Reed (Member # 137) on 03-06-2000, 05:55 PM:

Hey, ANYTHING'S better than those ridiculous CURVED splices Universal exchanges used to put in! They were terrrible... ALWAYS came apart.

Posted by John Pytlak (Member # 331) on 03-07-2000, 12:35 AM:

For continuous feed-on printers, the location that a lab splice falls in a reel is strictly a matter of chance. The printing negative (including leader) is run in a loop, so a splice in the leader can and does happen occasionally.

Trailers are often printed from looped negatives, or may be printed onto raw stock "short-ends" that have been spliced together.

Ultrasonic splices in raw stock may have undeveloped emulsion trapped in the weld, which is more unsightly and less reliable than the same splice on processed film. The splicer heats the film to make the weld, so there may also be some discoloration around the splice due to sensitization of the undeveloped emulsion. If the splice is not on the frameline, or is unsightly, remake it!

In making an ultrasonic splice, the splicer cuts the film so there is a slight overlap. The overlapped ends of the film are pressed into intimate contact with an ultrasonic transducer ("anvil" or "horn"), which causes the molecules in the thermoplastic polyester film support to vibrate so fast that it heats up and melts together, forming a weld. I recall the ultrasonic frequency is about 50 kilohertz.

Ultrasonic splicers do NOT work with nitrate or triacetate film, as they are not thermoplastics, and char rather than melt together to form a weld.