Restoring worn out DAT tapes

[Bob Grant]

The only surviving masters are the ones used to duplicate cassettes and they've been used a few times too many I think. In many, many places I'm getting what I assume are digital dropouts, the waveform just holds its last value for several milliseconds. There's a few other issues but those I can deal with but the dropouts are the show stoppers.
I'm finding that if I play these tapes back again the dropouts are occuring in slightly different places, so perhaps over several passes all the data could be recovered but even so patching all this back together sounds like a hideous amount of work, total length of the program is over 3 hours! I've got Vegas and Sound Forge, SF should be the best tool for sample accurate work, anyone care to suggest a better app for this kind of work.

Another question, I'm capturing this audio via SPDIF from a mid range Sony deck, would any of the higher end decks do a better job of recovering the data, could it simply be a problem with tracking, do some decks have wider latitude in the servos. Sorry if that's a dumb question, I've plenty of experience with DV where that can be the case.

[Steve House]

Are you capturing the audio from a DAT into a computer, dubbing to another digital recorder, or playing back through a digital mixer? First thing I'd check is to make sure that the target device is properly set up to sync its clock to that in the incoming SPDIF data stream.

It is true that magnetic media is more ephemeral that is generally thought. The substrate is stable, the magnetic material is stable, but it turns out the glues that hold the two together aren't. ....

[Jay Massengill]

As Steve said, definitely check that the two devices are locking together so the digital signal can be recorded cleanly. If that's the problem, you might be able to solve it with a few mouse clicks or switch settings.
I guess there's no chance to perform playback on the deck that originally recorded the tapes?... Otherwise it could be an alignment issue, or actual tape dropouts, or dirty heads on the playback deck.
Some decks will give a readout of error rates, but that wouldn't necessarily help you solve the problem if it is the tape itself, just give you a better indicator.

[Bob Grant]

Tapes are being played out of Sony DAT via optical SPDIF and into PC, PC is clocking off SPDIF clocks. I've previously transferred dozens of DATs without problem. Monitoring these tapes through headphone jack on DAT deck yields same results as captured file. The client sent these tapes to me in the hope that I could work some magic on them as they'd heard the same problems on their gear.
The tapes are labelled 'duplication master' and 'cassette 1 sides 1 and 2', in other words they've likely been played 1000s of times!
Looking at the waveforms is quite interesting, where the problems occur the waveform holds its value, nice straight line as a DC voltage, clearly it's a digital problem, if it was in the analogue domain the waveform would have some sag over time.

[Jay Massengill]

Those pesky clients, sending you tapes they already knew would take a miracle to fix... (Just for kicks have you tried capturing a known good tape to make sure all is still well with your system? Their old tapes could be clogging up your heads.)
You may have to follow your first idea of capturing multiple passes and putting it back together. How frequently do these random dropouts occur? I'm sure it would be a pain, but if it's important to the client you could charge accordingly for that kind of labor. Any chance one of those duplicated cassettes still exists? I know those would suck, but modern noise reduction techniques work pretty well. Too bad they can't locate an archived DAT or a data backup of the file. Good luck!

[Bill Binder]

Couple of things...

You might want to post your question over at TapersSection because there are a lot of folks over their that have been dealing with DAT for a long time, and some of them must have dealt with your same issue over the years -- trying to resurrect old DATs.

Another idea, which might be obvious considering your comments above, might be to download the DAT several times to different files. Then bring all of the versions into an NLE like Vegas, get them all lined up perfectly in sync, and take it from there. I could see just mixing them together, which might be better than doing nothing at all, or better yet, just cutting back and forth between tracks whereever you are having glitches (assuming that the glitches don't always happen in the same spots). That might not be that bad actually, kind of like a multicam edit in a way.

[Ty Ford]

[QUOTE=Bob Grant]
I'm finding that if I play these tapes back again the dropouts are occuring in slightly different places, so perhaps over several passes all the data could be recovered but even so patching all this back together sounds like a hideous amount of work, total length of the program is over 3 hours!
\

Maybe the deck you're using needs cleaning or alignment.

Regards,

Ty

[Glenn Davidson]

I have seen this problem many times. I have found that Sony machines and Panasonic machine do not always get along. When a problem like this happens on our sv-3800 machines I can usually get them to play in a Sony consumer machine or another Panasonic (I have 4).

I also found that wet cleaning the head drum with denatured alcohol and chamois swab works, also.

The panasonic sv-3800 has a feature that will display error rates so you can see the problem before it is audible.

With the Sony/Panasonic problem in mind, I always thought it kinda ironic that they interconnect via SPDIF ..Sony Panasonic Digital Interface.

good luck..I only had to do the splicing of good passes once.

[Jay Massengill]

The "P" was for Phillips, maybe that's why they don't get along.

[Stephen Finton]

Originally Posted by Jay Massengill

The "P" was for Phillips, maybe that's why they don't get along.

Ironically, SPDIF stands for SONY/Phillips Digital Interface Format. :)

EDIT:Just realized that was what you were referring to...