Re: Re: K5 distorted output from patches using Source-2

Hi,

some things that might help you:

Datasheet is here, no need to post it:
http://focus.ti.com/lit/ds/symlink/pcm55.pdf
Product Page: http://focus.ti.com/docs/prod/folders/print/pcm55.html

Texas Instruments still produces the DAC, on the product page you can
see that Arrow and Rochester Electronics have some in stock.

As you might have guessed, one DAC outputs S1 sounds, and the other outputs S2.

Cold solder joints are common in this synth. In the first place, I'll
suggest you to rework all the solders on the PSU regulator board. A
lot of them fail.

The noise comes from the demultiplexing switches, the thousands of
opamps placed after the DAC as well as the VCAs.

Here's the diagram (what I remember) :
DACs ->
demux switches (one chip per DAC divides the mono signal in 4
different ones) ->
Sample & Hold circuit ->
S1 CH1-4 & S2 CH1-4 individual mixing (using high value resistors to
prevent feedback from other channels in the S&H circuit, adds noise)
->
VCAs (specs show they're quite noisy)->
Opamp Buffers ->
CH1-4 outputs.

The MIX and headphones outputs are connected to the last buffer stage
outputs, mix the signal and go through one more buffer.

I was thinking of experimenting with the output stage by connecting
two PGA4311s right at the output of the S&H opamps, and mix S1&S2
after the PGAs to see if noise levels would improve. I just lack a bit
of time to build the required stuff to program a PIC to control the
PGA chips.

Your DAC might have a problem, but since there are a lot of external
adjustments, it would be a good idea to check the components around
the DAC first.

Cheers,
Antoine

2008/7/14 narfman96 <narfman96@...>:
> The balance only controls the volume for S1 and S2. If S1 is all the
> way down S2 is cranked. Other than the DDA envelope (oh yeah the DHG
> levels too) ahhh, where can S2 be turned down to sublight.
>
> The aftertouch was easy. There's 2 pots on the backside of the MI-004
> board they say Gain (VR1) and Offset (VR2) on them. Open Midi-Ox and
> watch the pressure change the values for aftertouch from 00 to F7.
> Adjust these pots to get the best performance for your preferences.
> Offset is where it starts to kick in. Then the gain adjusts how
> quickly F7 cranks in. I like a smooth contollable range that doesn't
> rip my tendons out.... That's why I can tell the notes from G3 to D4
> in this K5 aren't as responsive as the rest.
>
> Whoohoo!!! I'm talking to Kawai about the Service Manual. They are
> soooo gooood in this day and age of crapping on us hardware guys.
> Oops did I say e-mu? Awww darn! Dirty bird!! I want my money back!!!
>
> Anyway the manual should be on its way....  Fran
>
> --- In k5synth@..., James Bong <jamesbong <at> ...> wrote:
>>
>> How does it sound if you turn S1 all the way down, and
>> start turning up S2? Same distortion?
>>
>> Oh, and where was the after-touch adjustment? I'm
>> always afraid I'm going to tip my stand over when I'm
>> using after touch modulation!
>>
>> James
>> --- narfman96 <narfman96 <at> ...> wrote:
>>
>> > Yes I do have a scope I just haven't gotten that far
>> > yet. Yesterday I
>> > rebuilt the keyboard. This synth has a nice action
>> > when it's all
>> > greased up and triggers every note. I found the
>> > aftertouch
>> > adjustments and got it more sensitive than the key
>> > bending tipping
>> > over the stand mode it was originally set for. The
>> > aftertouch strip
>> > has a few spots that are less sensitive but hey it's
>> > a beast....
>> >
>> > If it will help I can post the DAC's datasheet here
>> > in the files
>> > section so you can see what weird output
>> > configurations it can have.
>> > Either current or voltage outputs can be used. There
>> > isn't any pot
>> > for the offset zero. Also it seems like Kawai
>> > doesn't use much of the
>> > gain that's available. Maybe that's why it's so
>> > noisy.
>> >
>> > The patches sound like they should if I turn the
>> > balance so S2 is off
>> > (-31). Except of course they're thin and lifeless
>> > without the other
>> > voice mixed in. I saved a bank that way so I could
>> > check out the
>> > keyboard. I can see where some of the noise comes
>> > from. The high
>> > frequency response goes way out there. Most of these
>> > 80's and 90's
>> > synths have low pass filters. This thing sounds wide
>> > open on top end!
>> > Probably so you can hear the 128 harmonics in all
>> > their glory.
>> >
>> > Hopefully I won't be beat when I get home from work
>> > so I can play
>> > around with this some more. Still haven't heard from
>> > Kawai yet about
>> > the schematics. Also they always have me sign their
>> > non-disclosure
>> > form so I'm leery about openly posting them in the
>> > group. Of course
>> > helpful individuals are a different story. Right???
>> >
>> > Fran
>> >
>> > --- In k5synth@..., James Bong
>> > <jamesbong <at> > wrote:
>> > >
>> > > You wouldn't happen to have an oscilloscope to
>> > verify
>> > > that the output of this chip is actual problem, do
>> > > you?
>> > >
>> > > If you do get the schematics, could you possibly
>> > get
>> > > them scanned to share with the rest of us?
>> > >
>> > > James
>> > >
>> > > --- narfman96 <narfman96 <at> > wrote:
>> > >
>> > > > I resoldered everything involved in this part of
>> > the
>> > > > circuit. There
>> > > > is a feedback resistor on the DAC for the audio
>> > > > output. This measures
>> > > > 6.8k same as the working one. Maybe tonight I'll
>> > try
>> > > > lowering this
>> > > > value to see if the gain decreases.
>> > > >
>> > > > I'm not having much luck finding a PCM55HP DAC.
>> > > > Digikey lists it for
>> > > > $38.50 with a long, long lead time... Crap!
>> > Anyone
>> > > > know of a source?
>> > > >
>> > > > Hopefully Kawai gets back to me today about the
>> > > > schematic.
>> > > >
>> > > > Thanks, Fran
>> > > >
>> > > > --- In k5synth@..., James Bong
>> > > > <jamesbong <at> > wrote:
>> > > > >
>> > > > > Does this chip have a bias circuit to set the
>> > > > min/max
>> > > > > output voltage scale? It could be something as
>> > > > simple
>> > > > > as a cold solder joint finally loosing contact
>> > and
>> > > > > changing the bias. I'd check/re-solder the
>> > solder
>> > > > > joints to everything connected to the chip,
>> > and
>> > > > the
>> > > > > chip it's self.
>> > > > >
>> > > > > James
>> > > >
>> > > >
>> > >
>> >
>> >
>> >
>>
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>

--

-- 
Antoine Deschênes

------------------------------------