Discussion about the Korg PolySix synthesizer

I found the data sheet.   I don't think I can attach it here.

The difference is the standby current.   The L part is less than 2 uA.
The non-part is less than 20 uA.   So it affects your battery life.
If you want to see where you stand, apply 3V in place of the battery
to your populated board.  It doesn't need to be in the PolySix.
Put an ammeter in series with whatever is supplying the 3V,
and see what the standby current of the SRAM is.   
Of course you don't want the battery connected when you do this 
test.  Maybe you installed a socket for it?   If you used a socket
for the SRAM, then you can test several parts and pick the lowest 
one.

If you want the data sheet, contact me using the contact info
at my site.   tauntek.com   And I will email it to you.

Bob Grieb

--------------------------------------------
On Sat, 2/6/16, jw_dewdney@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:

 Subject: [PolySix] POLYSIX SRAM CHIP REPLACEMENT?
 To: PolySix@yahoogroups.com
 Date: Saturday, February 6, 2016, 2:47 PM
 
 
  
 
 
 
   
 
 
     
       
       
       Hi - I'm just putting the finishing touches on
 one of Jed's 367 boards - and in my efforts to remove
 the TC5514 chip with solder wick - i'm a bit worried
 I've cooked the poor thing (though I haven't fired
 it up yet - who knows what the long term effect will be on
 the patches - i hear SRAM is particularly sensitive to lots
 of factors).
 
 
 At any rate - my question is this - I just
 bought some replacement TC5514 chips online - but didn't
 notice that I got the AP3 version instead of APL3... does
 anyone know the difference? I can't seem to find any
 data sheets online from a valid source (just the usual spam
 sites)...
 
 
 
     
      
 
     
     
 
 
 
 #yiv6460587043 #yiv6460587043 --

I uploaded the data sheet into the files section.   I misspoke about the standby current.
The L part is 0.2 uA versus 20 uA for the non L part.   You can still try the 
ammeter test.  Japanese mfrs tend to be very conservative with their
standby current numbers.  The parts you have might be good enough.
Up to you...


--------------------------------------------
On Sat, 2/6/16, jw_dewdney@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:

 Subject: [PolySix] Re: POLYSIX SRAM CHIP REPLACEMENT?
 To: PolySix@yahoogroups.com
 Date: Saturday, February 6, 2016, 3:32 PM
 
 
  
 
 
 
   
 
 
     
       
       
       Thanks Bob - no that's QUITE a convincing
 argument/explanation... I'll get the APLs instead...
 anyone had any experience with cooking their SRAM chips?? or
 are there any good tests to find out? I guess checking patch
 storage etc would be the obvious - but not sure if there are
 better/more critical tests?
 
 I was hoping chip quik would work for
 extracting these but it's useless without a decent
 extraction tool - which I'm trying to get my hands on...
 (open to suggestions on this too!)
 
     
      
 
     
     
 
 
 
 #yiv8765232817 #yiv8765232817 --

Please allow me throw in a few more words.

Reading the datasheet carefully you will find that the currents so far discussed are maximum currents - maximum in the sense of among a large group of samples. The typical current is 0.05uA, even for the non-'L' variants. The L variants seem to be just screened to meet lower maximum current specs.

These currents are listed Under conditions of VDD = 5V and Ta=85C

On the last page you can see how IDDS (standby current) goes up exponentially with temperature, from about 200nA at 85C down to just 4nA at room temperature (20C).

So you may actually have trouble to even measure these 4nA with hobbyist measurement Equipment.

With such current your 125mAh battery will last 3000 years ... theoretically. In reality the battery has it's own leakage current we don't know here.

So unless your ammeter can really measure something of uA's your safe to use even the non 'L' parts.

Ah, and another point touched before: the battery is only serving current to the SRAM and one more CMOS logic gate 14011, IC25. While the synth is not powerd, the other parts are separated from that supply through D15.

Cheers

FWIW, I've made pin-for-pin replacement of the RAM using regular low 
power RAM SMD chip and PCB adapter with pins. Only slightly wider than 
real chip. Still have a few if anyone's interested. I used to sell them 
to a tech repairing P6 by the dozens, and since dawn of KiwiSix there's 
no use for them anymore.

You can check RAM's leaking current in the circuit, inside P6 by 
measuring voltage across the diode that goes between battery and RAM 
with decoupling capacitors (C36+C40) and 4011 (IC25). It should be 
nearly zero. If it's more than 0.1V, leakage is too high. But you gotta 
be quick, because voltmeter input resistance will charge the capacitors 
pretty fast, unless you have an option to switch it to 1Gohm.
There is no such diode in service manual, but it has to be added when 
replacing NiCad for Li battery. That's the diode you put instead of 100 
ohm R91 resistor that was originally there.

Roman

W dniu 2016-02-06 o 21:32, jw_dewdney@yahoo.com [PolySix] pisze:

>
>
> Thanks Bob - no that's QUITE a convincing argument/explanation... I'll
> get the APLs instead... anyone had any experience with cooking their
> SRAM chips?? or are there any good tests to find out? I guess checking
> patch storage etc would be the obvious - but not sure if there are
> better/more critical tests?
>
> I was hoping chip quik would work for extracting these but it's useless
> without a decent extraction tool - which I'm trying to get my hands
> on... (open to suggestions on this too!)
>
>

yes, that was 32KB part, basicaly the smallest one you can get in 
reasonable prices now. In first version the PCB was ready for banking, 
few of address lines were connected to shift register in hope to create 
panel mount device to change banks. Later I skipped that, so new PCB is 
smaller. I never got back to the idea of bank switcher.

BTW, since you can get TC5514 from China now, this is probably 
irrelevant anyway.

Roman

W dniu 2016-02-09 o 15:22, grantbt@jps.net [PolySix] pisze:

>
>
> That's cool Roman.  I assume you can't buy the original SRAM in SMD, so
> it must e much larger. Any chance there is banking allowed for? ;-)
>
> GB
>
>