SV: Software Counter/Oscillascope

[*Star Project]

For replacing uPD444c memory
chips I would go with teh ones that Jameco
sells. they call them 6514 I believe but it
also says 444 on their site so. They are
cheap aswell so.
By the way f = 1/T were f = frequency & T = period time
so 1.8ms =~ 555.6Hz while 2.2us =~ 454.55 KHz
By the way does anyone now where to buy
samsung sram chips near sweden.
Im looking for the K6T0808C1D-DB55, K6T0808C1D-DB70,
KM62256DLP-5L or KM62256DLP-7L. The access tiem doesnt
really mather its the LL-pwr version that im after.
(data retention power is 0.2-3uA at 3V)
Its for a TB-303 memory replacement board
that hopefully will get a memory->PC interface PIC
later on.

/ Andreas H


-----Ursprungligt meddelande-----
Från: owner-synth-diy at node12b53.a2000.nl
[mailto:owner-synth-diy at node12b53.a2000.nl]För danial stocks
Skickat: den 9 januari 2001 12:58
Till: synth-diy at node12b53.a2000.nl
Ämne: Re: Software Counter/Oscillascope



Sorry folks been on hols, missed the start of this thread.. but anyway, mem
probs can be the mem chips themselves.. I've seen the upd 444c's blow out
before.. replacement with 2114 is pin compatible but draws too much current
for mem retention, there are a few obscure other ones which could be used
tho UM6104 can I think, altho I spliced in a 6116 when I had to fix this, it
actually fits one 6116 in the socket over the 2 memchips original spots.. it
hangs off one end, the unused datya lines, which you terminate in a res
network to gnd.. I think one other pin needs to be changed, but its a 1 chip
replace, no glue logic, and you get an extra mem bank as a bonus!..
Cheers,
Dan
> >I've been over the circuit board dozens of times checking 4 broken
> >traces and the such, tested 100's of parts, but with no avail.
> >The last thing that i feel sus about in this unit is maybe the CPU
> >clock and/or "INT Clock" (is this interupt clock?) is mis aligned.
> >The "Int clock" adjustment is made via trimmer TM1 while the CPU is
> >one of those tuning coil "thingie majigs".
> >I asked this question on AH with no responce about what the symtoms
> >of an mis alligned CPU clock and INT clock would be in an XOX
> >machine???
>
>I seem to recall that the int clock is being used to do the panel scanning.
>Every 1.8 mS, it tells the processor to try to look at the next group of
>buttons.
>
>If the clock were misaligned, there might be two different scenarios.  If
>the period were too long, it wouldn't be watching the buttons as often, so
>it might seem a little sluggish or unsreponsive when you push buttons...but
>I'd venture a guess that it's have to be significantly slower than 1.8 mS.
>
>If the clock were faster, yopu might find that buttons didn't register,
>because the scanner would move on before the contacts on one button were
>done bouncing.  I think that if the Int clock is sick, the switch scanning
>will be sick.
>
> >I've been working 4 a couple of weeks now on a non funtioning 606,
> >all the sound generators are fine, yet it refuses to write anything
> >to memory (playing previously stored patterns and tracks work
> >though).
> >this might not be the case as it has no problems playing back
> >previously stored patterns and tracks, just anything involving
> >writing to memory.
>
>Check the health of the "-WE" lines to the Ram chips.  When you're writing
>a
>pattern, those lines should be hopping up and down a bit...of course
>without
>a scope, they'll be very hard to see...those cycles are probably darned
>quick...guessing somewhere in the 1 or 2 uS range.
>
>Even if you don't have a scope, at least check the continuity between them
>and the glue logic that generates -WE...and then from the glue back to the
>processor.  It could be that the Glue is dead...assuming all of the above
>suspects are getting power.
>
>Byron Jacquot
>

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.