Film-Tech Forum ARCHIVE: Dolby Digital--Error Number Increase Cause

This is topic Dolby Digital--Error Number Increase Cause 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=008923

Posted by Bill Enos (Member # 440) on 09-17-2007, 12:20 PM:

What causes the error number on the processor to go up? We're playing a print of Sicko that is running at an unchanging 2. Examined in hand the DD track looks no better to the eye than prints that run at 5 or 6, definitely not unscuffed or undamaged, there are prints that look very good and run high numbers.
?????

Posted by Steve Guttag (Member # 268) on 09-17-2007, 01:23 PM:

Everything causes the number to go up...it basically represents how hard the processor is having to work to recover all of the information. If the print has too much azimuth error or the corners are not easily found...you name it.

Posted by Monte L Fullmer (Member # 2797) on 09-17-2007, 01:24 PM:

I call it a lab thingey. On trailer paks, it's interesting that each trailer can issue a different error number. Then, when the feature begins, it ususally locks into a certain number.

Posted by Brad Miller (Member # 2) on 09-17-2007, 02:05 PM:

The causes of Dolby Digital error rates increasing as you play the print are from dirt and scratches on the Dolby track itself. While scratches and scuffing are pretty obvious causes, few people seem to realize that the data blocks in between the perforations are extremely tiny, such that a given sized tiny speck of dirt in the picture area that most people wouldn't find distracting from the movie will block a chunk of the Dolby Digital data if that same tiny piece of dirt is in between the perforations.

Wet clean your prints regularly and make sure you don't have anything causing scratching/scuffing in between the perforations and the error rates will actually go DOWN. (This is assuming you do this as "preventive" measures and not "emergency recovery" measures, for by that point there isn't always a happy ending.)

Posted by Frank Dubrois (Member # 3042) on 09-17-2007, 03:23 PM:

We have a brand new Cat 701 reader and the error rates are fairly consistant at 4-5. Granted, this is in a smaller house and gets the prints that are about 1 week from leaving. We clean our prints everday with a film cleaner and filmguard. Shouldn't we get much lower error rates?

Posted by Louis Bornwasser (Member # 3063) on 09-17-2007, 03:38 PM:

Remember there is nothing to be gained by a "low number" as long as nothing is faulting. The only other thought is that you have a better chance if your alignment "is in the center of normal." That way some variation in the printing of the film can be accepted.

4-5 on a penthouse reader is OK. I have seen lower on Kelmar basement readers on Century projectors, including "0" on a direct drive model. I have not seen an increase of the numer with time. In other words, wear does not seem to be a factor. If it starts at 3-4, it will stay there for the run, even 4-5 months later. Louis

Posted by Brad Miller (Member # 2) on 09-17-2007, 05:30 PM:

Louis, on Century projectors, yes. On others, not so much.

As far as the cat701 reader, those have always been the shitty ones that Dolby made. Any of the other models will track better.

Posted by Tim Reed (Member # 137) on 09-17-2007, 06:09 PM:

Believe it or not, I have gotten zeroes after aligning two houses at GC's flagship theatre in MA... on Cinemeccanica V-5 basement readers! Sure surprised me. Staff said they'd never before seen zero on any of them.

Posted by Sam D. Chavez (Member # 1841) on 09-17-2007, 07:55 PM:

The V5 and V8 readers are among the best in this regard if you live through the clunky adjustments. The reason is the large diameter drum is very good on vertical movement or jitter. It is not so good on lateral movement, but the SR.D hardware/software is not so fussy about this.

I've seen error rates down to low 1's and 2's with either penthouse or Basement readers. There was a period of time where the 650 software gave optimistic low error readings so we need to factor in when we saw these numbers.

A really good tech with a well running projector can get great results from basement readers, but these are the exception rather than the rule as a lot of cinemas get zero attention until something breaks.

Posted by Tim Reed (Member # 137) on 09-17-2007, 08:17 PM:

FWIW, these were DA-20s and CP-65s.

Posted by Joe Redifer (Member # 3) on 09-17-2007, 08:35 PM:

quote: Louis Bornwasser
Remember there is nothing to be gained by a "low number" as long as nothing is faulting.

True, but I noticed that sometimes when the number changes to F that it does not switch back to analog, most likely because the F happens very quickly and goes away. However you can definitely hear it in the auditorium when the Dolby Digital is F'ing. It sounds like small clicks and weird audio gaps/interruptions. We had many Kelmar basement readers and they would F all the time, even if the numbers were 3 or lower. We'd see readings similar to 2-F-2-F-5-4-4-1-F-0-4, etc all in that order as the print played in the span of maybe 30 seconds in any given auditorium on any given print and it would never switch back to SR. Sometimes it would even F-dot. F'ing Dolby Digital.

Posted by Bastiaan Fleerkate (Member # 3826) on 09-23-2007, 09:45 PM:

We had a cat. 700 penthouse reader and a DA-20 with a CP-65 and we had some prints that would run 0's but there were also prints that were F'ing all the time, so that we had to switch to SR manually (turning the reader off) because the processor was switching from digital to analog and then back to digital multiple times a minute... the sound in the auditorium would 'skip' when the change to digital happened, but was always in sync...

Is it true the new red-light readers are performing better and generating lower error numbers? Or has it all to do with the type of Dolby Digital Processor you've got...

Posted by Steve Guttag (Member # 268) on 09-23-2007, 11:30 PM:

The red light readers are inherently inferior...even for digital sound scanning. A Cat 700 should out-track a Cat 701 or Cat 702. Furthermore, the Cat 700 has a user changable "EPT" lamp...the current LED readers require some skill in the LED replacement and an O'Scope for proper alignment. Note the BACP reader has a much more user friendly LED/head assy user changeout without test equipment.

Steve

Posted by Louis Bornwasser (Member # 3063) on 09-24-2007, 09:53 AM:

I've "saved back" a cat 700 that I use in critical rentals for just the reason Steve mentioned. Louis

Posted by Patrick de Groot (Member # 858) on 09-24-2007, 11:24 AM:

When you ittermittently see the F go check if it's the reader or the movie. It is annoying to hear bleeps and clicks all the way through the movie. The processor never converts to analog in this case because the F is short. It just plays corrupted digital audio data.

Posted by Mark Gulbrandsen (Member # 72) on 09-24-2007, 02:57 PM:

While I am too a CAT 700 fan beware that there are no light pipes available for them any longer. If you're moving them around pack them in an exceptional foam lined case! If you crack it or scratch its surface its an expensive upgrade at Dolby to make it into a CAT 702, basically about 2/3 the cost of a new reader and not worth it. We had one such reader like this that a customer bought used off of E-Bay. The surface of the light pipe was scratched and although we found an optical person at Hill Air-Force base that polished the scratch out for us there was still a crack within the light pipe that prevented normal illumination of the track. For this reader I machined my own LED mount half to accept Dolby's standard LED and the reader is working fine again.

Its somewhat surprising that Dolby has not gone to a single super bright LED as was used on the Sony 3000 penthouse. All of those Sony LED's around here are still going strong after many many years of service. At least Sony did one aspect of the SDDS system quite well.

Mark

Posted by System Notices (Member # 2357) on 12-31-2009, 09:06 PM:

It has been 829 days since the last post.

Posted by Ron Curran (Member # 3631) on 12-31-2009, 09:06 PM:

Sorry to drag up this old thread but it is for good news.
I saw my first 0 (zero) on my Dolby Digital processor today. Old system, new print of The Princess and the Frog.
Some printers can do it right.

Posted by Jeremy Weigel (Member # 4208) on 12-31-2009, 10:50 PM:

We had a trailer not too long ago that was giving solid zeros and occasional ones. I took pics of it, because I'd never seen it happen before.

Posted by John Wilson (Member # 269) on 01-03-2010, 04:11 PM:

quote: Ron Curran
new print of The Princess and the Frog.

Did you get a print from Thailand too, Ron? [thumbsup]

Posted by Chris Slycord (Member # 4239) on 01-03-2010, 05:36 PM:

A bit off-topic but I remember opening up a CP-500 to check the error levels since I didn't know why it'd reverted to SR.

Instead, the thing said "O - H - E - L - P". It made me chuckle that the device was asking for help. Thankfully, a tech fixed it the next day.

Posted by Ron Curran (Member # 3631) on 01-03-2010, 06:45 PM:

John,
Can't find a source on the box or print - just Technicolor and Ship to Australia, so we can assume it came from out there.
Image is nice too. Just when I thought we needed a new lens, along comes a print that looks great.

Posted by Jeremy M Smith (Member # 4913) on 01-09-2010, 02:34 AM:

Wish I could be so lucky I have one that does the 4 - F - 5 - F ect fixed it once with changing the bearing in the layon roller what a mission 6 months later it back to it's old tricks. One day I might see a 0

Posted by Demetris Thoupis (Member # 833) on 01-09-2010, 02:57 AM:

Jeremy what projector are u using?

Posted by Jeremy M Smith (Member # 4913) on 01-09-2010, 03:22 AM:

Demetris it is a simplex 5 star sound head

Posted by Demetris Thoupis (Member # 833) on 01-09-2010, 01:18 PM:

Hello Jeremy,
What is the status of the dampen roller? Is it steady or is it bouncing back and forth during playback? Is the lateral adjustment tight or is it moving? Is the sounddrum spinning properly or is it glitched? When you turn it by hand it should spin freely on the flywheel and take a while to stop. If its spinning with difficulty then that is where your problem is. Check whether the flywheel has come loose for some reason. Also confirm the correct video level (4V) is coming in your processor.
Demetris

Posted by Jeremy M Smith (Member # 4913) on 01-10-2010, 02:54 AM:

What do you mean by the dampen roller is ti the one that lays on the top to hold the film in place for the sound drum ?

Posted by Demetris Thoupis (Member # 833) on 01-10-2010, 03:00 AM:

There should be a roller after the soundrum which acts like a balancer and acts like a spring when you try to move it.
Demetris

Posted by Ron Curran (Member # 3631) on 01-28-2010, 11:17 PM:

Just to update. Our Princess and the Frog print was still bringing up 0 error rates at the end of its 4 week run.
FilmGuard used every alternate session.

Posted by Monte L Fullmer (Member # 2797) on 01-29-2010, 12:16 AM:

quote: Demetris Thoupis
There should be a roller after the soundrum which acts like a balancer

..Called the "snubber roller" and a lot of times, the goop that adds the return friction under that spring goes bad and have to be re-gooped so the snubbing action can return to balance out the film flutter as it goes around the scanner drum.

Posted by Damien Taylor (Member # 4255) on 01-29-2010, 08:35 AM:

Is this the same as the oil damped rollers on the old optical soundheads?

Posted by Monte L Fullmer (Member # 2797) on 01-29-2010, 12:49 PM:

Yep, the same .. anything to "snub" out film flutter and to have constant film speed over the scanner drum...

-Monte

Posted by Ron Curran (Member # 3631) on 02-12-2010, 02:05 AM:

New print of Valentine's Day showing constant Fs. No dts discs so running SR. Other prints running on same machine perfectly with low error rates.

Posted by John Wilson (Member # 269) on 02-12-2010, 02:42 AM:

Is that the whole print Ron or just certain reels. It'd be odd to be the whole print. Mine's doing solid 4's.

Sorry, I don't have discs for it either.

Posted by Monte L Fullmer (Member # 2797) on 02-12-2010, 02:47 AM:

Have a print of "2012" at one of our discount houses that goes from "5" on the trailers to straight "F" on the feature.

..yep, SRD track is lightly scratched. I plugged in the WinDRAS from my laptop from the encoder (DA-20) and definitely a loss of "corners" comes up on the display.

Posted by Ron Curran (Member # 3631) on 02-12-2010, 07:23 PM:

John,
At least the first 3 spools. One of the later spools is showing 7s and Fs but SR all the way is a safer bet. The time code looks a bit off centre so discs probably wouldn't help.

While on the subject of dts, we had 3 prints in a row that wouldn't play and we were just aboutto blow our month's profit on a new reader when the next print ran perfectly and so did every dts print since.

This is the most trouble we have had with dts since the Jurassic age.

Serves us right for staying with film instead of dci.

Posted by John Wilson (Member # 269) on 02-15-2010, 04:37 PM:

Ron,

Have you checked on some discs from Roadshow?

Posted by Ron Curran (Member # 3631) on 02-15-2010, 06:53 PM:

Strange thing. After several runs spools 3 onwards ran in Dolby Digital. Now spool 2 is running fine. Spool 1 stubbornly staying with Fs.

Surely could not be film cleaner; this is a new print.

Posted by Demetris Thoupis (Member # 833) on 02-16-2010, 02:16 AM:

Ron if it is a da20 or cp500 I would try changing cat 670 video acquisition board.
Demetis

Posted by Eric Robinson (Member # 2966) on 09-21-2010, 02:28 PM:

Anyone know how to perform horizontal adjustment "lateral tracking" of red LED for CAT701? Or do you just screw it on and its perfect?

Posted by Louis Bornwasser (Member # 3063) on 09-21-2010, 09:22 PM:

Full alignment is possible for the 700 series of readers. What you want is a combination of "lateral guide" and camera alignment. Louis

Posted by Eric Robinson (Member # 2966) on 09-21-2010, 11:47 PM:

Is there a document available which describes the method of these alignments?

Posted by Monte L Fullmer (Member # 2797) on 09-22-2010, 01:01 AM:

I've used WinDRAS to perform the horizontal alignments so the "box" image is in the middle of the window.

..along with reading the lateral position and the azimuth adjustments, then let the scope do the finess stuff.

When it comes to use test film. cut a hunk of film from a new trailer to make a large loop that rests on the upper guidance roller to lace through the entire machine and soundhead and around the front of the machine back up to the upper guidance roller.

-Monte

Posted by Sam D. Chavez (Member # 1841) on 09-22-2010, 09:45 AM:

The screws holes that hold the CCD board in place are generally slotted to permit some lateral movement.

There is no lateral adjustment on the roller that hold the film over the sound drum. However, the flanged guide roller does wear out and this can have the effect of moving the film out of position. This needs to be checked first before trying to align with DRAS.

There is also an alignment film (Cat. No. 566?) that should be used with DRAS to get the film into the center of the CCD. This is important as each film soundtrack is off by a little laterally relative to each other. The SR.D system can read film +/- .006 inches off in either direction, but only if you are in the center in the first place.

Posted by Demetris Thoupis (Member # 833) on 09-22-2010, 10:00 AM:

I always use a Cat 69T to callibrate with the osciloscope and then a precalibrated Cat 530 test film to tweeq final adjustments. I have though referenced one Cat 69T loop to various precalibrated readers and use that as well and I never had any issues so far although the appropriate film to do it is the Cat 530.

Demetris

Posted by Louis Bornwasser (Member # 3063) on 09-22-2010, 07:49 PM:

We recently made tooling to locate the center of the lens barrel EXACTLY in the center of the DD checkerboard. Interestingly we found that 98% are "off" horizontally, probably from the factory. We then could move the sprocket and shim the lateral guide rollers to agree. This allows the camera to be centered exactly in its barrel.

You cannot imagine how much signal we now get compared to before.

btw: if anyone needs a 700 (light bulb) converted to 701 (LED), we now do that conversion. Louis

Posted by Steve Guttag (Member # 268) on 09-22-2010, 08:52 PM:

Out of curiosity...why would anyone want to convert from an EPT lamp to an LED? The lamp is user changable and cheaper.

Posted by Eric Robinson (Member # 2966) on 09-26-2010, 05:27 PM:

Our 700 using the EPT lamp has a beautifully flat line across the top of the envelope (on the scope display.) I thought it to be much better than the choppy line you get with the LEDs.

Anyway, my original question must not have been too clear. Is there a document published by Dolby Labs. which specifically describes the mechanical alignment adjustments for the LED or lateral guide roller? Or, is this information confidential? Or, is it just supposed to work perfectly when you put it together. According to Louis - there is some error involved with new factory units.

Thanks for all of the input otherwise.

Posted by Steve Guttag (Member # 268) on 09-26-2010, 05:33 PM:

I believe in the Cat 701/702 manuals there is a discussion on LED/CCD alignment. Also, check the DA20 installation manual...I seem to recall a section there.

We have a bit of a cottage business R&Ring all sorts of penthouse readers that have been all sorts of abused/used. Also found that there was a quite a few of the Dolby penthouses where they didn't get the sprocket aligned with the lateral guide rollers.

For the record, we have never converted a white light (Cat 699/Cat700) to a red light reader. And yes, the white light readers do a better job of tracking than red.

Steve

Posted by Eric Robinson (Member # 2966) on 09-26-2010, 05:52 PM:

Thank you Steve [Wink]

I found the info I was looking for on pages 94 through 97 of the DA20 manual on the film-tech website. Just what I was looking for

[thumbsup]

Posted by Louis Bornwasser (Member # 3063) on 09-26-2010, 06:13 PM:

FYI, Steve, all of the Dolby readers, 699/700/701/702 are essentially the same. I have converted ONE 700 to 701 because Dolby could not supply a light pipe.

And, yes, I agree with you, the bulb is a better deal for all. Louis

Posted by Steve Guttag (Member # 268) on 09-26-2010, 06:50 PM:

The 699 was the most different of that group.