[sdiy] Juno 60 woes

[Jack Jackson]

Hey, thanks for the reply.

I added the reset circuit as you suggested, but still no luck. I've checked the keyboard matrix scanning lines (going into the 74LS42 from the CPU) and only some of them are active, suggesting the CPU is not behaving properly. The interrupt (INT) line on the CPU (coming from the patch board CPU) is constantly at 0V and it is active low. So i think some input to the CPU is not behaving properly and causing it to get locked in this state.

Any more ideas guys?

Cheers,
Jack

----------------------------------------
> From: jackdamery at hotmail.co.uk
> To: rsdio at audiobanshee.com
> Subject: RE: [sdiy] Juno 60 woes
> Date: Wed, 3 Jun 2015 20:23:31 +0100
>
> Hey, thanks for the reply.
>
> I added the reset circuit as you suggested, but still no luck. I've checked the keyboard matrix scanning lines (going into the 74LS42 from the CPU) and only some of them are active, suggesting the CPU is not behaving properly. The interrupt (INT) line on the CPU (coming from the patch board CPU) is constantly at 0V and it is active low. So i think some input to the CPU is not behaving properly and causing it to get locked in this state.
>
> Any more ideas guys?
>
> Cheers,
> Jack
>
> ----------------------------------------
>> Subject: Re: [sdiy] Juno 60 woes
>> From: rsdio at audiobanshee.com
>> Date: Wed, 3 Jun 2015 09:17:51 -0700
>> CC: synth-diy at dropmix.xs4all.nl
>> To: jackdamery at hotmail.co.uk
>>
>>
>> On Jun 3, 2015, at 5:16 AM, Jack Jackson <jackdamery at hotmail.co.uk> wrote:
>>> So I've ordered a new CPU for my non working Juno 60. Apparently I should perform the modification to the reset circuit (add diode and change C133 to 10uF). Is this critical to the operation of the CPU. In the service manual it says it just makes sure the lowest C key is functional.
>> I'm fairly certain that two separate issues are involved.
>>
>> First, a bug related to Transpose and Release was found in the original CPU firmware (-238), and this was fixed by releasing a second CPU with different firmware (-380). That's the low C key fix.
>>
>> Second, when they released the new firmware, they must have used a different revision of the uPD8049C. Apparently, the new part has more requirements for the reset circuit, and thus, changing the CPU requires modifying the reset circuit.
>>
>> In other words, the reset circuit doesn't fix the low C key error, or else they would have just modified that circuit and left the original CPU. The reset circuit is only to allow the newer CPU chip to function properly.
>>
>> In general, processors can be very picky about the Reset input. I had an early PIC PCB where I could wave my hand over the board without touching anything and it would reset. That's because I got the reset circuit wrong on the first revision. Each processor has different requirements - some require nothing at all on the reset pin (which is why I assumed I had the right circuit, based on earlier PIC chips), while others require a fairly specific circuit. In this case, it looks like the diode allows the cap to charge up quickly to engage reset a.s.a.p. on power-off, while the original resistor allows the cap to discharge more slowly to disengage reset after a slight RC delay after power-up.
>>
>> Anyway, if your new CPU is the uPD8049C-380, then you should confirm that the diode is there. If you're replacing the CPU with the same model that was already in your Juno 60, then you shouldn't have to change anything unless someone forgot to update the reset circuit.
>>
>> I'm not certain, but I would assume there's no harm in having the diode even if you're installing the uPD8049C-238, but as I said above, chips can be very touchy about reset.
>>
>> Brian
>>
>