Tech Flashback (& Experiment): Fujifilm DLTtape IV Cartridge

Thanks to one of my friends at the university, while I was out of action due to my ankle, he managed to save me quite a few “old” things that were thrown out. One of them was this – a Fujifilm DLTtapeIV Cartridge.

DSC_8678 DSC_8679

Digital Linear Tape, or DLT for short, was a backup tape format first developed by Digital Equipment Corporation and later bought by Quantum. This system used a half-inch tape inside a cartridge with a single spool, with the take-up reel being inside the drive itself. The leader is specially formed so that the drive can “catch” the beginning of the tape, and thus, the tapes cannot be ejected without first being rewound. This format evolved through multiple generations, and all of the technical details are nicely gathered in the Quantum DLTtape Handbook (well worth a read, at least the first 87 pages or so).

In many ways, this format pioneered the reliability which was a hallmark of simple mechanisms, minimal handling guides, and fixed head linear recording, and was very much emulated by the open, less-controlled, Linear Tape Open (LTO) format.

Such tapes don’t normally land in my hands as many home users didn’t resort to tape backup, and corporations usually carefully dispose of their tapes. In this case, I managed to get several “still new, unused” condition tapes. The tapes themselves were probably fairly pricey and dates back to around 1994-1999.

Unfortunately, I didn’t have any drive to use them with, but it’s worth a close look and a teardown.

DSC_8680 DSC_8681 DSC_8682

The cartridge comes in a translucent plastic box, with two large rectangular raised parts which engage with corresponding depressions in the bottom of the next box for stackability. The box opens on the left edge to reveal the cartridge. The cover itself has a lip along the edge, likely for dust-control.

The front cover itself gives a good amount of information about the cassette itself.

Fuji-DLTtapeIV-Carton-Outer

The cassette media itself uses “ATOMM” technology, which implies it is a Metal Particle type layer. The cassette itself was Made in Japan – there were only a handful of manufacturers at the time. The tape is claimed to be 12.65mm, with a capacity of 40/70/80Gb depending on which DLT drive is in use (due to different recording formats). All of these capacities are claimed at a 2:1 compression ratio, the native capacity is just 20/35/40Gb.

Fuji-DLTtapeIV-Carton-Inner

The inside features strict information about storage and use of the write protect tab. On the right side, are peel off perforated card segments which can be written on and slipped in a corresponding notch in the cartridge for labelling purposes.

The cartridge itself, as characteristic of the DLTtapeIV format, is in a chocolate brown colour. It’s also roughly square in shape.

DSC_8683 DSC_8684

There are no clear windows – you know the tape has to be fully wound on the cartridge if it’s ejected! The bottom reel drive has a very interesting shape, which has many many points of engagement. There’s a hole at the bottom which one of the reel lock disengagement points.

DSC_8685

DSC_8686

The top of the cartridge has another hole which corresponds to the second reel lock disengagement point. The left part has a pull-down latch which opens the tape flap door. Visible is the leader tail which is used by the drive to engage with the tape. The reel lock teeth is just visible through the slot at the bottom.

DSC_8688 DSC_8689

The bottom edge of the cartridge has the labelling area – the piece of cardboard can be inserted through the slot and slid into the rails. A captive write protect mechanism slider is visible, when the orange is visible, the cartridge is write protected. This also controls an indent on the right side of the cartridge for write-protection detection, as can be seen below.

DSC_8687

The final side of the cartridge is the left side – it’s not really interesting …

DSC_8690

Teardown

Since I had several cartridges, I didn’t mind sacrificing one in the quest of knowledge. Luckily, as the cartridge was held together by screws, opening it is as simple as undoing the Phillips head screws on the bottom. The cartridge practically springs apart.

DSC_8691

The top of the cartridge is seen on the left side. The door and the two spring-loaded reel lock mechanisms are visible. The spool is on the right side, with the finely-teethed reel lock plate visible. The spool itself is sprung against the top lid with a circular spring, which rides on a precision smooth-running bearing (likely, ball bearing, similar to that on computer fans).

DSC_8692

The tape is protected by a specially formed leader – this is the initial segment. The leader is specified to be 18 inches in length, and is enough to fully wrap around the tape (thus, keeping dust and environmental impact upon the thin tape to a minimum). The leader itself is quite thick and sturdy compared to the metal particle tape.

DSC_8693

Here, we can see the leader attach to the metal particle tape with splicing tape. This implies that the recording surface is actually wound recording side face down which is quite smart!

Experiment

Well, looking at this can you think of any other common half-inch tape format?

I’ll give you a moment …

and

… if you’re anything like me, at this point, it’s where you go eureka! Of course! VHS!

The first thing I was thinking of was “how much tape is on a DLTtape IV cartridge?” which was very quickly answered by the DLT Handbook linked earlier – 1800 feet. That’s a lot of tape. In fact, a PAL E-300 5-hour cassette only has 1427 feet of tape. Visually judging, the tape seems to be spooled on a reel with a larger hub, so it’s got to be very very thin tape.

The next thing I asked myself was “how good of a VHS tape would DLTtape IV metal particle tape be?” Initially, I thought it might be good because of the high quality of tape required for high data densities, but I also realized it would probably not work well since metal particle tape can be more abrasive to heads, and helical scan recording causes high tape-to-head speeds. The other problem was that metal tape has higher coercivity than chrome dioxide and ferric formulation cassettes, which might mean the VHS deck may not have a strong enough magnetization to record to the tape properly.

The only way to know for sure, is to try it! Don’t do this at home – you could cause yourself some damage to your VHS deck if you don’t do this right!

So, how does one wind a DLT tape to a VHS spool? Well, first you grab a VHS you don’t care about (in my case, a spare Q-stores E60 cassette with large spools, makes winding a bit quicker although you can’t wind quite as much) and clip off the tape, leaving leader intact. Then you “splice” together (with sticky tape, as I didn’t have splicing tape) the tape to the leader … and begin winding …

DSC_8695

Yep. By hand. But uh-oh … the DLT handbook warned me of this …

DSC_8697

That’s the beginning of tape “hole”. This hole indicates the beginning of the real recording – the prior tape is used for calibration and directory recording. So then, I disposed the tape and then respliced after this hole. For reference, there’s an end of tape hole which is exactly the same, but is 610cm from the end of the tape. The tape isn’t actually held to the DLT hub by anything – it’s just wrapped on top of itself and held by tension.

After a few hours of winding, I got this …

DSC_8698

DSC_8699… a full spool. It’s actually quite a lot of tape because it’s about half the DLT reel …

I eventually threw out the remaining tape as I unwound it to find out what’s at the end (yeah, that wasn’t worth the effort).

 

 

 

When spliced back to the other VHS reel, and placed into a cassete shell, it turned out to be an E-209! That’s right – just one minute short of three and a half hours on reels which normally hold no more than an hour and a half of VHS tape. That shows just how “thin” and fragile DLTtape IV is.

DSC_8705

Speaking of fragile … the thinness of the tape causes it to curl around the edges, and that’s pretty normal of most thin metal particle tapes. It’s also exactly shiny, and seems very smooth, but how it treats the heads of my VHS deck is another story … the tape wasn’t designed for helical scan recording, so it might not work very well. Maybe it’s too thin to achieve reliable head-to-tape contact, aside from the coercivity issue.

DSC_8701

The answer: as expected, it didn’t work so well. The playback of a static menu from a DVD recorded to the tape was very noisy chroma wise and had many dropouts. The noise in the chroma, and the entire chroma dropout during tracking suggests the RF signal level on the tape was very low, indicating it wasn’t recorded well into the tape. At least, now we know …

dlt-vhs

So, uh, don’t use DLTtape IV as a VHS tape? *chuckles to self*

About lui_gough

I'm a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!
This entry was posted in Computing, Tech Flashback and tagged , , , , , , . Bookmark the permalink.

10 Responses to Tech Flashback (& Experiment): Fujifilm DLTtape IV Cartridge

  1. Darkstar says:

    Heh, I know these (still have a few tapes and a DLT8000 drive in one of my computers). But trying to use it as VHS tape really never came to my mind. Are you sure you spooled it the right way though? I think the recording layer is on the outside for both VHS and DLT (as can be seen in the PDF manual you linked to), but on your image it looks like you wind it so that the outside ends up on the inside of the VHS cartridge?

    • lui_gough says:

      I’m not sure that the record layer is on the outside for DLT. The reason I wound it the other way is because I presumed the DLT guys wound it with the data layer facing into the hub (opposite to VHS) as the splicing tape that connects the leader tape to the DLT tape was visible on the outside. As I would assume (and I believe rightly so) the splicing tape always is stuck on the back face of the tape. Judging from the way the tape curls “away” is another hint that I likely have it the right way around. Maybe I can get a knife and scratch at it just to double check …

      – Gough

    • lui_gough says:

      In fact, your idea is so intriguing, I had to just try it (and managed to snap the tape in the process). Unfortunately, my knife didn’t scrape anything off either side, so I resorted to my “other trick”. I stuck a clear adhesive tape to the top and peeled it off quickly (it tore the tape and left some remnants of magnetic material) and stuck this to white paper. Then I repeated it for the bottom side of the tape.

      From the image, you can see that there’s magnetic material stripes left on the top tape, but not the bottom. The bottom managed to take some material from my desk – that’s the brown stuff.

      So I think this means that I’ve spooled it the right way around …

      – Gough

    • lui_gough says:

      Maybe if you look at the DLT Handbook figure 5-3 on page 59, you will see that it implies that it is wrapped with the data side inwards when it’s in the cartridge itself. The rollers only contact the rear surface of the tape (a reliability measure), and note the way the tape comes out of the cartridge.

      Ah well. I think I expected the result that I got anyway – it’s a high density, high quality, high coercivity tape which is difficult to record to. Maybe if the field current in the VHS heads are increased, it would be possible to record a very “sturdy” reliable signal which won’t degrade much over time (compared to lower coercivity media). Maybe the result would be different for Super VHS (which is more demanding on media and may have used a higher coercivity media).

      – Gough

  2. Darkstar says:

    Yeah, you’re right, I was going by the pictures in the PDF alone, and there it looks like the spool comes out the cartridge the other way round. Checking with an actual cartridge makes it clear that the data is actually on the inside

  3. sparcie says:

    I wouldn’t have thought of doing that! I was interested to find out that DLT and LTO tapes are linear as opposed to helical, but it certainly makes sense as the drives wouldn’t have room for the rotating drum and mechanism. As you say there may also be a reliability factor as well. It’s been quite a while since I’ve had an LTO drive apart!

    Cheers Sparcie

  4. Aardvark says:

    DEC was Digital Equipment Corporation in Maynard, Massachusetts.

    I used those types of cartridges dating back to the first ones marketed by DEC as the TK50 and later followed on by the TK70. They were slow as hell but the capacity was quite astounding for their size. At my present employer, I still have a 40GB DELL DLT drive and quite a few of those FUJI tapes as well as Compaq branded ones. We used to use them for archival purpose and offsite storage (they still work) but we found it was cheaper and faster to go out buy 64GB thumb drives instead.

    • lui_gough says:

      Thanks for that – I was doing too many things at once and managed to mangle DEC up. Surprised to hear that some are still in use given the age and speed – but I suppose their long term reliability is proven which is good for archival. I think the main deterrent would be the proprietary nature of the format (meaning less drives to choose from compared to LTO) and the age might be an issue if the drive(s) ever fail.

      – Gough

  5. Bruno Lori says:

    Try a S-VHS recorder instead of a VHS drive. Standard VHS uses Chrome tape, S-VHS uses metal particle as DLT does.

    • lui_gough says:

      Great idea, however, sadly, I never owned an S-VHS deck nor cassette and they were not that popular overall. Their expectation and use of higher energy/coercivity tape will give a much better chance of working – but the idea was mainly borne out of curiosity, rather than a serious attempt at re-use. I’ll keep that in mind if I ever have the chance to repeat the experiment.

      – Gough

Error: Comment is Missing!