|DA-88 Overview and FAQ
My Name is Eddie Ciletti. Before moving to West St Paul, Minnesota in May 1999, I was an authorized Tascam service center in New York City for eleven years. I was the first to service the DA-88. At least fifty percent of my monthly column in EQ Magazine is devoted to the topic of digital tape machines. As of April 2000, I now write for MIX Magazine.
WHAT IS THE DTRS FORMAT?
"DTRS" is Tascam's name for their eight-channel digital recording format using Hi8 tape. The models are: DA-88, DA-98, DA-38 and now, the DA-78 (24 bit).
AVOID DOWNTIME: Know the Facts
Head life concerns ALL users and models, but more so for the
DA-88 because it was the first DTRS machine. Hard
to believe that I still service many DA-88's that have never been updated.
To start, the DA-88 has a fan for its LINEAR power supply while the DA-38 and DA-98 have a more efficient SWTICHING supply that requires no additional cooling. The stock DA-88 fan pulls air through the unit, from the sides as well as the tape insertion slot, which brings in dirt that can contaminate the clutches and back-tension felts.
DA-88 head life is based on environmental factors both on the user end as well as the manufacturing end. While Tascam has not ever implemented my fan mod, I feel it is essential and, if you don't believe my words, then believe the images below!
The Reel Tables
The Reel Table Assembly has been redesigned three times. It consists of seven parts: the color-coded "key" that makes contact with the cassette hub, a circular magnet, a large plastic spacer, an anodized circular steel plate attached to a white gear base, a black plastic optical encoder and small plastic spacer. The magnet, large plastic spacer and steel creates a clutch, the tension of which is determined by spacer thickness. Tascam only specifies the tension range, not the dimensional characterisitcs of the spacers. The assembly can not be purchased as individual parts.
Reel table height is also not specified by Tascam, but I always treat the supply and take-up reels differently. There is a split washer that holds each reel table in place. Normally, a 0.1-inch washer is under each optical encoder. On the take-up side, my tweak is to put a 0.10-inch washer under the split washer and a 0.005-inch washer under the optical encoder. (this lowers the Take-up Reel Table so that the tape does not have to be forced down so hard by the upper flange of the adjustable guide just prior to the cassette shell re-entry.) On the supply side, I put a no washers under the split washer and two 0.1-inch washers uner the optical encoder. (This raises the Supply Reel Table so that there is less drag on the lower flange of the Impedance Roller.
REEL (Tension) PROBLEMS
When the Supply Reel clutch becomes contaminated, the normal tension (20 cm/grams) in Reverse-Cue can be double to 40 cm/gramms, the resulting increase in tape tension seriously decreases the life of the heads. When combined with a typically high factory back-tension setting, head life falls far short of the expected 1,000 hours. It should be twice that, and I believe the machines I service have a much better chance of reaching the 2,000 hour mark.
A failing take-up clutch will have erratic tension putting a heavy strain on the capstan servo system causing the machine to take longer to lock to itself, let alone another machine.
To summarize, it is TAPE TENSION -- and not any particular brand of tape -- that is, in my humble opinion, primarily responsible for the most common form of premature head wear.
The importance of learning more about the machines we use can not be
understated. I don't expect users to become service technicians,
but it is to your advantage to become
familiar with the subtle ways machines communicate their problems to us.
Respond by investigating Error Rate, Listening to the Mechanism, Popping
the Cover and most important, schedule routine maintenance, like,
uh, every 250 hours. OK? Click here for
AVOID THE FLAME
Following "The Checklist," are the Frequently Asked Questions typical of those found on the NET concerning the DA-88. While users typically lay the "flame" either the tape or Tascam, I can only say, "Paranoia will get you nowhere. Knowledge is power!" Sure, Tascam may not be the most forthcoming. I know TEAC Japan creates documents that address problems and their solutions, but US service stations have not been consistently kept up to date. I am working with Tascam's customer service department on these issues.
I believe service facilities and users will find my documentation most helpful. It is not my policy to trash manufacturers. I may be unhappy with their "performance," but in too many cases the humans I interact with are not to blame. Whether purposefully or simply by being understaffed, the end result is that barely enough information is made available.
Since premature head failure can be costly, here's a check list to keep you sane.
Each number is a link to a more detailed explanation.
1.) Use the TEST mode to check the error rate (using VU meters 1 & 2),
2.) Use "DTRS" tape which can be Sony, BASF, Ampex / Quantegy, HHB or Apogee. MAXELL does not have the DTRS logo, but it is good tape.
3.) Do not record anything important on the first or last five minutes of tape. While new heads have no problem with the extreme "head" or "tail" of the tape, you can extend the usefulness of worn heads by avoiding areas that have a potentially higher error rate.
4.) REVERSE the fan and add the filter. The mechanism should then be removed and thoroughly cleaned.
5.) To minimize head wear, CHECK the tension spring position. The position of the tension arm position MUST be set to the longest tape available. Then the tension can be measured to ensure that it is within the window of safety.
6.) Learn to manually clean the heads and to query the machine for total number of cleanings via tape.
7.) Listen to the machine closely in REVERSE CUE mode. Is there an "odd" repetitive sound? If so the supply reel table / clutch is bad and is putting excessive tension across the heads. Replace the Reel Tables.
Without disclosing more than I am legally able -- in such a public forum -- all DTRS labeled tapes give good performance. I am surprised when people recommend Metal Evaporated (ME) tape. It does have high output and low error rate but, at least earlier in its history, short term performance. (I personally think ME is better suited for archival / minimal shuttle useage, but perhaps this too is a function of tension...)
Tapes with the DTRS logo include: AMPEX, BASF, HHB, APOGEE and SONY. If you think you are having problems with any manufacturer's tape, try another so long as it's got the DTRS logo. The exception is Maxell, but this is also good tape. TDK and FUJI MP formulations, however, are not suitable for stock (early and unmodified) DA-88s. The record bias and an ATF resistor must be changed (on some machines) to improve performance with these tapes.
Manufacturers of all sorts of products have good days and bad. Mass production does just that -- but quantity is not quality. This applies to tape and machines. I know that tape mechanisms outside the "window of tolerance" make it to the marketplace because to analyze and pull them from the production line requires human intervention and that costs money. When new, a digital tape transport can meet the most essential specificaton, error rate, but when aged, the inherent flaws become obstacles that are difficult to tame.
Techincians are encouraged to replace entire mechanisms rather that determine the problem, which is sometimes the design itself. For example, many tape transports have too few adjustments, amking them simpler to align when all parts are perfect and impossible to align when out of tolerance. For example, only Panasonic uses a machined, cast-metal transport base. All the others are stamped sheet metal. Panasonic allows height adjustment of all precision guides yet no machine provides "tilt" or "zenith" adjustments -- although some "V" blocks can be shimmed to improve the RF envelope. In short, it is not possible to compensate for all mechanical tolerance variations.
The best we can all hope for is to take advantage of the affordability of both the products and the tape stock. Own spare machines, make safeties on a regular basis and learn to read and promptly act on what minimal feedback -- in terms of an error message or error rate display -- is available. I highly recommend that all users contact manufacturers and demand an ERROR RATE BUTTON and not some cryptic combination of keystrokes. In fact, ERROR RATE should always be displayed and a warning buzzer sounded whenever the threshold is crossed for an extended period of time. Once the Error Indicator is lit, it is too late!
JUST THE FAQS
1.Timecode will "stutter" in fast wind mode
A QUESTION, A REPLY and then a COMMENT by Eddie Ciletti
>At TWI we are experiencing severe head wear problems which 'allegedly'
may possibly be due to our use of Maxell p5-60HMbq professional Hi8 tapes.
Initial feedback from Tascam appears we may not be alone.
Same tapes here, but 90 not 60. Advised and supplied by HHB after the machine was serviced and upgraded by them. No problems so far. What do Tascam say is the preferred tape now? I changed to the Maxell after drop-out problems with several other makes - I understood Maxell to be the Tascam recommendation at the time. Can't fault the reliability but no way of knowing how the head wear is progressing.
EDDIE CILETTI COMMENTS
Tascam (and my self as well) will always suggest changing tapes when users have problems because all manufacturers will occassionally have a bad batch and, if no improvement is observed, the tape can then be ruled out as the cause. In addition, if the tension arm
Dennis Weinreich asks:
We are getting through an incredible amount of heads, 1 set per month approx....
and yes, Tascam are aware of this and are being careful not to say anything.
EDDIE's REPLY: "Paranoia will get you nowhere. Knowledge is power!"
On a more serious note and perhaps more to the point, Tascam has always been very accomodating if it has been determined that the heads or any other part of the machine have prematurely failed.
I did just recently (12/97) see two machines with "B" record head problems. The first time this problem was encountered, I replaced the head assembly. The second time, I noticed that the error rate could be improved by adjusting the PB EQ -- which told me the data was readable. I reset the EQ to the test tape and tweaked the record current down. This made the head useable so I'll be monitoring the progress. Tascam did change the record current way back between lot 10 (sn 100000 and heads numbered under 20) and lot 20 (200000 heads numbered 20 and above) machines. Both machines were of a similar lot, had just under 300 hours with between 7 and 9 head cleanings.
AVOID TENSION HEADACHES
DA-88 head life is shortened by excess back tension. Tension is set by sliding a plate (in between the two reel tables) to which is attached a "long" black coiled spring. New heads will perform well with minimum tension. For old heads, it may be too late. Tension "servo" calibration involves setting the tension arm so that it wraps the tape around an impedance roller. This adjustment must not be used to set tension. An 8mm tension tape is available from both Tascam and Sony. I believe the price is about $150. Supply tension should be about 8 cm/grams and not exceed 10 cm/grams.
Because take-up tension is applied via magnetic clutches (the reel tables) and back-tension via felt, tension is affected by accumulated dirt and humidity.
Projected head life is 1,000 hours. I have seen machines with more and less. The sooner you know what the tension is -- and have it corrected, if necessary -- the longer the head life will be. The machine keeps a log of its total head hours as well as total shuttle time and number of number of times the cleaning tape was inserted. Users should also keep a log. Battery failure will destroy the log. Always check the battery voltage as well as look for leakage (on the top most SYSTEM board) as this will cause other very unusual problems.
Dirty heads are not the only problem. Cleaning the rabbet is also important but this is not a user-serviceable item. Erratic Take-up tension can also make a tape hard to track. It should be between 10 gm/cm an 15 gm/cm and not vary more than 2 gm /cm either way. Dirt and lint on the precision guides is the equivalent of "flutter" and is also a deterrent to good tracking and low error rate.
Tascam made a few changes to the transport. The most obvious was the addition of a tape cleaning mechanism. This "self-cleaning kit" removes particles from the tape in the forward direction only. It also includes a roller that is supposed to clean the head. I always install the former, but never the latter as it gets in the way of the right post roller set screw. The cleaning kit is replaced free of charge by authorized servicers.
The reel table clutches were changed as was the slide cam, sector gear and its cam. There was also a clip added in early models to stabilize the back tension felt. The heads were changed fairly early on, I believe somewhere between serial numbers 100000 and 200000. A "newer" head has a "20" or higher number stamped on top. There will also be a red mark on the head amplifier. Older servo boards have two trim pots, later servo boards have one.
Serial number range
under 100000 (a really old machine that is likely
to need all of the above upgrades)
Four final points:
1.) Tapes made on marginal machines should be cloned.
2.) There are four heads: two record and two playback.
3.) It really helps to have access to an oscilloscope.
4.) Have a good relationship with a local service technician.
The DA-88 and MIDI