Tech: Akai 635D Line switch

Discussion in 'Audio Equipment' started by Pinsgonewild, Mar 1, 2012.

  1. Pinsgonewild

    Pinsgonewild Guest

    Hello I jumped over from the Pinball group hence the name:) I have my
    nice Akai 635D torn down to fix what I thought is a problem with my
    Line/Source Switch. It's not acting like a toggle switch. You have to
    either push it in or pull it out. Is this normal? I'm thinking not. It
    works fine but I thought it should toggle and lock between the two
    states. Any insites would be apprecieted.

    Regards Don
    Pinsgonewild, Mar 1, 2012
    1. Advertisements

  2. Pinsgonewild

    Johny B Good Guest

    Nice machine. I've had to google for images and, from what I can see, it
    looks like a standard push button toggle switch.

    My best guess is that the spring that pushes the button out has broken or
    come adrift.

    BTW, googling for images of that deck of yours was a bit of a nostalgia
    trip. I still own three Akai reel to reel decks, a vintage M8 (valves and
    the famous crossfield biassing system - it's classed as a portable tape
    recorder, in spite of its 47 Lb weight, simply because of its hard cover
    and stout carrying handle!), a GX630DB and a GX747.

    The GX747 had been bought on the proceeds of the insurance claim for the
    theft of the GX630DB which, long story, I managed to acquire a year or two
    later via a chance meeting with a customer who thought he recognised it
    from my description as being in the hands of an acquaintance of his.
    Needless to say, when I finally clapped eyes on it, I quite happily
    stumped up the 100 quid asking price (I'd already had the insurance money
    and wasn't about to try and undo the numerous transactions that had ensued
    as a result of the original theft).

    I was able to instantly identify the deck since I'd added additional
    recording calibration knobs on the back panel (sensitivity - critical for
    Dolby noise reduction, Bias and HF EQ). I'd invested a lot of time in
    undoing Akai's "Schoolboy Howlers" regarding some very strange cost
    cutting design choices and the replay and record heads alignment, so was
    thankful to have the chance to 'buy it back'.

    As was my practice at the time, whenever I purchased something as complex
    as a tape deck, I would purchase a full service manual from the
    manufacturer whilst the kit was still current.

    In the case of the 630 and 747 decks, I had no problems getting the
    service manuals, unlike the M8 which had required me to apply to The Rank
    Organisation (Akai's UK agent at the time) for a manual - in this case,
    due to difficulties in obtaining a fresh copy from Akai, they sent me a
    photocopy of one of their own workshop service manuals to tide me over
    until the promised and paid for manual arrived in stock. It never did and
    they sent me another photocopied manual many months later as a final
    settlement of the transaction. I could hardly complain since the M8 was a
    vintage machine even back then.

    I'm glad I did invest in the service manuals since I was able to repair
    the fullwave rectifier bridge cct used by the direct drive capstain servo
    controller in the GX630DB... twice, IIRC. I'd bought the manuals mainly to
    be able to correct any stupid design errors and improve on any substandard
    modules. Although this applied to the GX747, the only stupid mistake was
    in the plastic 'modesty' cover for the heads just catching the edge of the
    tape, requiring nothing more high tech to fix than a craft knife (there
    weren't any dolby boards for Akai's penny pinching buggeration to exhibit
    itself as a replay clipping issue mentioned in a magazine review).

    Anyhow, all that aside, I hope you're able to sort out that switch. The
    GX630DB uses a lever switch to select source or tape so doesn't present
    the same issues of reduced reliability[1] that seem to plague spring
    loaded pushbutton toggle switches. Sensibly (as in the 'common sense'
    meaning) the only such pushbuttons are on the Left/Right/Stereo recording
    mode switch (basically, a switch that might be exercised about once or
    twice a decade), the Dolby NR button and the LN/WR tape type selector,
    again, functions that don't get switched very often (unlike the
    source/tape switch which can see extensive use)

    [1] This switch did eventually start to show signs of intermittent high
    resistance contact so I simply used it to control a couple of sealed
    contact relays to do the actual switching of the signal paths. The high
    resistance issue of switch contacts doesn't present any problem when
    switching some 10 to 20 volts at 20 to 100mA. It's only an issue with
    signals that vary over the range of micro to milli volts and nano to micro
    amps. The thinnest film of tarnish (or silicone based furniture spray
    applied polish) is quite capable of stopping such low power signals in
    their tracks.
    Johny B Good, Mar 2, 2012
    1. Advertisements

  3. Pinsgonewild

    Johny B Good Guest

    That's a nice demonstration of extreme serendipity. ;-) Hell! Such a
    story involving the more common HiFi cassette deck models that were
    available at that time would be almost as serendipitous.

    BTW, did you actually have a workshop manual for the 360 deck or was
    using your own as a template for the rebuild out of necessity rather than
    an irresistable[1] 'showboating' exercise?

    [1] Given the same circumstances, I think I might have done the same,
    workshop manual or not.
    Johny B Good, Mar 7, 2012
  4. Pinsgonewild

    John Guest

    You forgot to include the all important silicon/germanium point
    contact diodes used as detectors in the radar recievers. ;-)

    regards, John (first posting using freeAgent rather than Opera's mail
    client so haven't figured out the sig block yet)
    John, Mar 9, 2012
  5. Pinsgonewild

    Johny B Good Guest

    repeat posting with FreeAgent and persona configured with sig block?

    You forgot to include the all important silicon/germanium point
    contact diodes used as detectors in the radar recievers. ;-)

    Regards, John
    Johny B Good, Mar 9, 2012
  6. Pinsgonewild

    Johny B Good Guest

    I'm pretty certain that the first _reliable_ solid state device was
    the humble point contact silicon/germanium diode which provided a much
    more sensitive detector device (at least in airborne) radar sets, than
    the vacuum tube rectifier diodes of the day. Modern (relatively
    speaking) examples, at least of the germanium varieties, would be the
    OA47 and the OA91 diodes.

    Sadly, for those of a more inquisitive mind, the logisitics of
    today's modern electronic devices, mean that a whole subsystem (which
    might even be the entirety of a small gadget) now occuppies the niche
    that a single discrete transistor on a 1970s circuit board used to
    inhabit. That's just simply the 'price of progress'.

    That price of progress, it seems to me, has reduced the numbers of
    individuals in the world's population (note: not a percentage of the 7
    billion world population) who can understand how the circuitry
    actually functions at the component level in both analogue and digital

    Even back in the 70s, very few individuals could have examined Akai's
    version of the otherwise 'Bog Standard' Dolby noise reduction boards
    and notice the cost cutting removal of a resistor and capacitor which
    had been the root cause of the replay clipping at a mere +8dbmW in the
    line output of the GX630DB as described in a magazine review (but they
    failed to take note of the similarly induced record clipping in the
    bass end of the spectrum).

    I decided to reverse engineer the additional components required to
    reset the bias level back to the mid voltage point without modifying
    the gain. I also went one better by replacing the emitter follower
    resistor with a constant current source that allowed it to drive loads
    as low as 3K ohms before assymetric clipping would become an issue.

    The net result was an increase of the replay clipping level by a
    further 10db (output clipping level at a more bombproof +18dbmW)
    ensuring that the replay of tapes recorded at excessive levels did not
    have to suffer the avoidable insult of additional distortion over and
    above that already present.

    A year or two after completing this corrective measure in all four
    dolby boards (two for replay and two for record), I finally got to see
    the original dolby circuit in a Wirless World magazine article and was
    suitably smug to see that I'd virtually matched the original
    components that Akai's design engineers had been forced to leave out
    of the circuit in order to save a small fraction of a percent on the
    manufacturing costs by the bean counters.

    I can't believe the engineers would have done this without making
    making a very pointed argument as to why such a compromise would be a
    "Very Bad Idea"(tm). It's just possible that they may have been
    persuaded that their customers would be careful enough to avoid high
    levels in the interest of minimizing tape distortion and therefore
    wouldn't require a replay clipping level beyond the +8dbmW mark but I
    suspect they'd simply been railroaded into accepting the cost cutting
    exercise thrust upon them from on high by the "Company Suits".

    Examination of the record/replay circuits in the GX747 revealed no
    obvious compromises (no noise reduction processing circuits) so I
    didn't feel impelled to make any 'improvements'. The only thing I had
    to fix was a purely mechanical issue of trimming some plastic off the
    head cover to eliminate interference with the tape. A rather serious
    "Schoolboy Howler"(tm) by the 'Cosmetics Design Team', thankfully, one
    that was quite trivial to fix.
    Johny B Good, Mar 11, 2012
  7. Pinsgonewild

    ~misfit~ Guest

    Somewhere on teh intarwebs Bill Graham wrote:
    Is that "real to real" as opposed to digital to digital?

    "Humans will have advanced a long, long, way when religious belief has a
    cozy little classification in the DSM."
    David Melville (in r.a.s.f1)
    ~misfit~, Mar 14, 2012
  8. Pinsgonewild

    ~misfit~ Guest

    Somewhere on teh intarwebs Bill Graham wrote:
    Heh! So I was right, "real to real" was a Freudian slip. I should have read
    further before replying. ;-)

    "Humans will have advanced a long, long, way when religious belief has a
    cozy little classification in the DSM."
    David Melville (in r.a.s.f1)
    ~misfit~, Mar 14, 2012
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.