Dolby Digital fallback

[Sigurd Wik]

Hey,

I got an issue with fallback to analog SR when running SR-D tracks, and wonder if someone could weigh in?

Setup: 2 projector changeover setup. Cat.702 heads > Dolby DA20 > Sony DFP3000.

Problem: Fallback occurs only in changeover, but fairly random. Sometimes fallback, sometimes not. Sometimes when changing over to to proj 1, sometimes to proj 2. I have pretty good and stable signal on the DA20, just did a cat.702 calibration with a scope and signal strength is solid from 4 to 5 on most prints. Put in a replacement DA20 with new cards. Behaves pretty much the same, never looses signal towards end of reel, but the DFP3000 falls back anyway. Falls back to SR for a couple of seconds, then pops back into SR-D.

Sigurd

[Marco Giustini]

I am not too familiar with a DA20+DFP3000 but I believe the DA20 should be wired to the automation port and tell the DFP3000 when to drop to analogue. So the issue is from the DA20 and not the Sony.

Now, many years ago, SRD used to regularly drop at reel change for a second. Dolby then released an update which came with the prints themselves - not a system update but an SRD subsystem update I think - and the issue disappeared as the SRD subsystem was able to cache a few frames around the splice so a splice or a changeover would not cause a dropout.

How did you wire your Motor Start connector at the back?

[Sigurd Wik]

Thanks! Not my install, but will have a check on the motor start connector when I'm back Monday.

You know when this print fix happened? We get prints in from the whole range of SRD history, so could test.

[Marco Giustini]

my CP500 stopped reverting to SR at reel change around year 2000

[Sigurd Wik]

Think I can rule that out then, we get the error on prints way later than that.

[Marco Giustini]

the way it worked was that Dolby released an update THROUGH the prints. Once the payload was read from the stream (on some non-audio bits) and checksummed, it would be applied to the board and stored for future use. So once the relevant HW was updated, it would work with any prints. I'm just mentioning that in case somehow that DA20 was unused for a long time and missed the update? I believe the "update" was always present in newer prints, the processor would just ignore it if not required.

What does the error rate display show at reel change? Does it say F or 0 (zero)? If memory serves before the update a reel change would result in a brief "F" with a dropout to SR. After the update, the error rate would read zero for a moment (I'd imagine it's because the buffered data is perfect, no need for error correction). Mind you, this is an old memory, might be hallucinating

[Sigurd Wik]

Interesting way of updating a processor. We have run a bunch of prints on the DA20, so imagine it must have caught the update.

Signal quality on DA20 is good all the time, like 4 to 5. No loss of signal at the moment of fallback.

[Christos Gartaganis]

Make sure that the motor start inputs on the DA20 are connected to relay contacts coming from both projectors and are activated the exact time you start each projector's motor when changing reels. This is crucial when using penthouse readers such as the cat. 702

[Stefan Scholz]

Set the DA 20 to digital only or however it was called. There may be some distorted short noise on XO, but people are used to that these days from their cellphone communications.
I gave up wiring fallback even at the time with SCPC, as it was more obstructive than the eventual short digital artefacts.

[Marco Giustini]

Signal quality on DA20 is good all the time, like 4 to 5. No loss of signal at the moment of fallback.

If you don't see F on the error rate display when the processor drops to SR, then it shouldn't drop to SR The DA20 would send the processor an automation command to go to SR after several milliseconds of "F". F can "flash" briefly and yet the DA20 won't tell the Sony to revert. You might hear the noise Stefan mentions (which I cannot tolerate!) but won't drop to SR.

If you don't see F at all, then something's happening. Check the motor start first.

[Pete Naples]

Interesting way of updating a processor. We have run a bunch of prints on the DA20, so imagine it must have caught the update.

Signal quality on DA20 is good all the time, like 4 to 5. No loss of signal at the moment of fallback.

What Dolby called 'Dynamic Updating'. IIRC the print carrying the update had to run three times before it applied.

[Sigurd Wik]

Set the DA 20 to digital only or however it was called. There may be some distorted short noise on XO, but people are used to that these days from their cellphone communications.
I gave up wiring fallback even at the time with SCPC, as it was more obstructive than the eventual short digital artefacts.

Do I understand you correct that you got this issue on your setup, and dropped the fallback function altogether to get rid of the problem? What is SCPC?

[Sigurd Wik]

Christos, Marco,

Just to be clear, motor start signal is present. It is my understanding that without the motor start signal present, our cat.702 setup wouldn't work on any changeovers, as the heads wouldn't start the necessary reading before changeover. But some changeovers are fine, some not.

[Sigurd Wik]

What Dolby called 'Dynamic Updating'. IIRC the print carrying the update had to run three times before it applied.

Oh wow. We never run a print three times, but would guess this would have been picked up in the commercial cinema it was installed a long time ago.

[Marco Giustini]

Christos, Marco,

Just to be clear, motor start signal is present. It is my understanding that without the motor start signal present, our cat.702 setup wouldn't work on any changeovers, as the heads wouldn't start the necessary reading before changeover. But some changeovers are fine, some not.

How is your Motor Start connector wired at the back of the DA20?