DIY solutions for the PS-M1 TASCAM power supply(new)

Help Support GroupDIY:

radcomtech

Member
Joined
Aug 11, 2017
Messages
6
I posted in another thread and I have four updates since 2017.

I originally built a number of adapting regulators to allow generic common power supplies to function
as the original TASCAM PS-M1 did.
The only difficulty is in finding the vintage original 4 pin trapezoid power connector.
I found a workaround in the use of older MOLEX female pins like the kind used inside PC Computers
in the 1990s; the four pin internal Hard Disk Power couplings which are still on modern internal
Desktop and Tower ACDC Power Supplies. They are about 10thou too big but a small section of
soft small solder wire will tighten them to use as connections to the TASCAM power Jack port
without hacking up the TASCAM in sacrilege!
Recall that the PS-M1 supplies +10Volt, -10Volt and "7VoltDC". The last rail is *not correctly labelled* from TEAC: That connection is actually a very low 7VAC supply that sends timing of 100 Hertz (Europe) to 120 Hertz(N.America) into various TASCAM equipment. The 644 required this timing from "mains" power to bring up the midicenter, group levels, and tape transport including the cool 'jog' thing.
It also allowed the Cassette to record/play twice as fast at 3-3/4 (3.75 IPS) with DBX.

I discovered that:
(1) The original PS-M1 power bricks were designed for a number of different models.
(2) MM1 mixers do not need that 100-120Hz timing signal marked 7VDC(incorrecly)
(3) 644 MidiCenter with 4 track Direct drive Cassette Tape draws quite a bit more juice off the +10Volt rail. Almost 1Amp at FFw or REV
(4) the -10V rail needs to be very stable with noise filtration constants to attenuate well above 66KHz to avoid contamination of all channel reproduction.

Therefore, the design of my adapter changed.
I would try to find a Power brick old school style that has +5, +12V(1amp) and -12V (at least 300mA).
I would ignore the +5Volt rail and run only one LED indicator and drop resistor from that line.
I would employ two linear regulators LM317 and LM337(negative version) with potentiometers to exactly vary the outputs to 10.0V and -10.0 Volts. And this worked 100%.
Then , I take the 10V rail output, and power the Astable Oscillator set for 120Hertz and produce a 10Vpp signal. This also worked.
I discovered the TASCAM equipment power pins inside the port are 1.5mm slightly smaller than the 1.85mm female molex pins with the workaround mentioned above.

I found a surplus "brick" style ACDC with "triple play" output; these bricks were commonplace for Backup Tape drives in the 1995 era like Colorado Tape 8mm and so on. Bonus; almost all have DIN5! plugs, so I used DIN5 right angle on my adapter board. This also worked well.

Finally, the last client with 644 has 3D Printer capability and at some point He may be able to model the 9x11mm trapezoid pattern to embed four Molex female KK series pins into a PS-M1 "plug".

I think I may go ahead and create another run of boards in February 2021 because I had received at least 5 or so inquires on missing or lost PS-M1 adapters.

I am also looking into the voltage adaption of the Chinese single input switcher (common) module, to modify it to supply +10 and -10 by change of just two resistors; in theory, for now. Then I would add a 7555 astable oscillator circuit to generate the 120Hertz frequency square wave to drive the more intelligent boards.

I will be attaching images from this new build that I have recently completed, using perf board.
I will contact all the back-email requests and letting them know my Kit price and timetables.

I find the TASCAM equipment fascinating and I have been at Electronics Repair since 1976 (student) and certified since 1979.
The attached image ; on the left is the +10V adjustable regulator, and the -10 is on the right. In the middle is the 555 Oscillator , and the DIN5 input connector is on that side of the perf board.
 

Attachments

  • IMG_20210128_225301105_BURST000_COVER_TOPsmall.jpg
    IMG_20210128_225301105_BURST000_COVER_TOPsmall.jpg
    69.1 KB · Views: 38

mjrippe

Well-known member
Joined
Dec 27, 2011
Messages
2,698
Location
Brooklyn, NY
Very nice work!  Those PSUs always seem to be missing.  I changed the plug on my MM1 to a military type I had in my drawer and built the PSU myself.  Wish I had known about your efforts  ;D

Now I've got other issues with the mixer though and it has been sitting for a year or more.  Do you happen to have the MM1 service manual or schematics?
 

radcomtech

Member
Joined
Aug 11, 2017
Messages
6
Very nice work! Those PSUs always seem to be missing. I changed the plug on my MM1 to a military type I had in my drawer and built the PSU myself. Wish I had known about your efforts ;D

Now I've got other issues with the mixer though and it has been sitting for a year or more. Do you happen to have the MM1 service manual or schematics?

I did return the schematics back to the client without scanning as the client was the Government sponsored CBC. I may see those TEAC MM1 mixer panels back again, soon and I may scan them. I also thought I found the same late 1970's document online, with block schema on pg 31
but you need an account (free?) for this link
TASCAM MM-1 OWNER'S MANUAL Pdf Download | ManualsLib

sorry for late reply

If you use your substituted bipolar supply for 644, do not ignore the 7Volt AC 120 Hertz requirement as some TEAC vintage products need that.
The 555 timer in astable oscillator mode is ideal for the task and directly feeds the 644 Tape drive so that it can index and it can JOG with timescale.

If the PS-M1 is used for the 644 deck instead of the MM1 vintage mixer, make *sure* that the 10Vdc is spot-on for both rails and that when the deck
is in fast-forward, the supply can produce the 800mA needed off the position 10 rail.

The 7VAC requirement did not seem necessary for the MM1; the original design used DC and this is _wrong_. For the 644, it heats up that input logic stage.

The 120Hertz originally was derived from AC Mains frequency, then half wave rectified and squared (easy to do with separate transformer winding.
In fact, there was a HAMMOND AC transformer that had triple 12+12+6.3VAC anti-phase (older terminology) outputs from 120/240 dual input windings . That 6.3VAC output could be used as the 7V square-wave source. The TEACs only needed 10mA or less of this signal, and the 644 requires 100 to 120 Hertz to operate. At 120 Hertz, the time indexes were spot-on. So you could fast forward to 5:33 every time and meet the same spot in the song. Although be warned that I had non-metallic tape and I did observe minor tape stretch effects on older tapes.

Its fun for about 60 seconds to listen to Led Zep and watch the cassette spindles go round in PLAY on tapes recorded for my Car Cassette in 1980, When I had Hair, Women and a Sports Car. Now I realize, jeez, this sounds like sh!t when compared with FLAC free lossless audio codec
playback.
I had about 60 seconds of nostalgia when staring at the rotation of the spindles, playing with the jog, and so on.
 
Last edited:

radcomtech

Member
Joined
Aug 11, 2017
Messages
6
I am building a new series of boards, 644 and MM1 and any other TEAC PS-M1 gear compliant,
with plenty of +10VDC rail current and a 100-120 Hertz 7V output signal.
No, I won't post the new design(s).
I am waiting for parts again, due at the start of summer.
radcomtech(at)outlook(dot)com.
The $120 USD kit will only include round pins and NOT the unobtanium 4pin trapezoidal shaped plug.
Only Canada and USA deliveries $20US postal shipping with higher costs for the UK and EU
once I have 4 Kits working with a _single_ 12Vdc input (2.1/5.5mm coax plug, + in center)

at this time I have one bench soldered reference kit working but I need that to service my Ottawa TEAC clients;
the image is in the first post- sorry, its not for sale(I am also repairing ), and it cost me the whole day to tune successfully;
but its based on linear 10V 7810's and the +10 one gets HOT in heatsink if a 644 REW or FF thru most of one side.
The heatsink gets into the 60's C for components that can take 100 C but I did almost scald meself trying to take
my entire 100+ cassette library to SIDE A / INDEX 0...
 
Last edited:

mjrippe

Well-known member
Joined
Dec 27, 2011
Messages
2,698
Location
Brooklyn, NY
Thank you, I have the owner's manual but it does not include schematics. I am only powering the MM1, so no worries about 7vac.
 

radcomtech

Member
Joined
Aug 11, 2017
Messages
6
I have built kits of switch-mode based PS-M1. Wide DC input. Small case. with no thermal heat build up. Adjustable. Able to handle 15 watts per rail and 644 tape-jog current surges. International AC operation compliant.

I had to swivel away from linear regulation due to the thermal issues and finding AC Mains adapters at around 16Vdc with the proper current. I found a China based solution off the shelf including a 555 based generator to create 7VAC that works and is set to 120Hertz at about 50% duty that runs from the +10Vdc rail. The board is modified with two trimmer pots to be adjustable [image2].

I tested with resistive loads for 7days with a recording Multimeter and the +10 and -10VDC outputs are stable. I wired four cables and terminated the cables with round female socket connectors that will fit right over the T644PortaStudio power port. Noise under full load is in the 750KHz to 1MHz range and I modified the outputs to suppress that noise to ~400mV; in tests, I could not hear any ripple noise at (644) outputs. Test loading was two 50 Watt 8 Ohm ceramic load resistors. Input bench supply was set at 14V, consistently drawing about 20 Watts.

Attached find new images.
Image 1: round socket-pins are repurposed from a older Floppy Molex connectors with Heat shrink. (Green is the 120Hertz 7VAC , Black=Gnd, Red=+10V, Violet=-10Vdc) *User/Clients will be responsible for getting the the four correct sockets into that trapezoid PS-M1 legacy connector at the rear of the 644 in correct order. See my earlier posted image for the connection adaptation. Also show are the two stripped ends connections from the 12V 3A adapter ; I will be locating 15Vdc 3A adapters because the inrush to the dual bipolar switch mode conversion is about 30 Watts.
Image2: both boards in lunch snack case with snap lid; right side board is 555 based 120Hertz time base generator
Image3: ACDC wall wart Mains adapter International 100-240VAC input with N.A. IEC Type A (adaptable for European 2pin Type C)
image4: two kits in small containers with lids closed

Sorry for the COVID related delays; the kits arrived in 12weeks from Guangdong 9 weeks.

I will be replying to all KIT requests soon this November. (cost TBD by DDP shipping and mains options)

Thank you, I have the owner's manual but it does not include schematics. I am only powering the MM1, so no worries about 7vac.
The MM1 did not ignore the 7V(ac) pin when it was *disconnected* from the PS-M1 port in 2019 tests; the panel digits and some functions froze.
 

Attachments

  • IMG_20211101_215033114.jpg
    IMG_20211101_215033114.jpg
    204.8 KB · Views: 6
  • IMG_20211101_215021241.jpg
    IMG_20211101_215021241.jpg
    202.6 KB · Views: 6
  • IMG_20211101_214915779.jpg
    IMG_20211101_214915779.jpg
    113 KB · Views: 6
  • IMG_20211101_214908257.jpg
    IMG_20211101_214908257.jpg
    257.9 KB · Views: 6
Last edited:

Latest posts

Top